工作流 /
图像 /
MiniMax H3 Image Generate + Edit | 16 MP, 8 Pictures
MiniMax H3 Image Generate + Edit | 16 MP, 8 Pictures
MiniMax H3 Image Generate + Edit | 16 MP, 8 Pictures
Make a 16 megapixel picture from a prompt, or edit a picture in plain words. For an edit, you can give MiniMax H3 Image up to seven more pictures and tell it what to take from each one.How to use itDrop the workflow into…
图像
38 个节点 1,425 次使用 AusBoss 285 KB
需要的权重
这个工作流引用了 1 个权重文件，其中 0 个能在本站模型库里找到。
MiniMax-H3-Image Generate + Edit (AusBoss).json
模型库暂未收录
节点构成
AUSBOSS_NODES_Callout
× 7
AUSBOSS_NODES_WorkflowSwitches
× 3
AUSBOSS_NODES_CalloutTile
× 3
AUSBOSS_NODES_WorkflowNoteTile
× 3
CLIPLoader
× 2
VAELoader
× 2
UNETLoader
× 2
AUSBOSS_NODES_Compare
× 2
AUSBOSS_NODES_MathExpression
× 2
AUSBOSS_NODES_WorkflowNote
× 1
AUSBOSS_NODES_Text
× 1
AUSBOSS_NODES_ImageCropRotatePad
× 1
AUSBOSS_NODES_Seed
× 1
AUSBOSS_NODES_Resolution
× 1
763be216-d720-52d4-b604-469834ebd74c
× 1
AUSBOSS_NODES_ShowText
× 1
官方标注的模型
minimax_h3_image_pruned_int8_convrot.safetensorsminimax_h3_image_pruned_w4a8.safetensorsminimax_h3_image_vae_fp16.safetensorsminimax_h3_pdmd_2nfe_comfyui.safetensorsqwen3.5_9b_w4a8_gptq.safetensorsqwen3vl_32b_minimax_h3_nvfp4_awq.safetensorsseedvr2_3b_int8_convrot.safetensorsseedvr2_ema_vae_fp16.safetensors
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "61bfdef2-c9d9-4033-bb77-4e1de15f3ff1",
3 "revision": 0,
4 "last_node_id": 470,
5 "last_link_id": 1132,
6 "nodes": [
7 {
8 "id": 1,
9 "type": "AUSBOSS_NODES_WorkflowNote",
10 "pos": [
11 -120,
12 30
13 ],
14 "size": [
15 560,
16 1340
17 ],
18 "flags": {
19 "pinned": true
20 },
21 "order": 1,
22 "mode": 0,
23 "inputs": [
24 {
25 "localized_name": "note",
26 "name": "note",
27 "type": "STRING",
28 "widget": {
29 "name": "note"
30 },
31 "link": null
32 }
33 ],
34 "outputs": [],
35 "title": "🆎 AusBoss",
36 "properties": {
37 "Node name for S&R": "AUSBOSS_NODES_WorkflowNote",
38 "cnr_id": "ausboss-nodes",
39 "ver": "2.15.0"
40 },
41 "widgets_values": [
42 "{\"v\": 1, \"title\": \"MiniMax H3 Image · Guide + models\", \"subtitle\": \"Make a 16 megapixel picture from a few words, or change a picture you load, with MiniMax-H3-Image. Give it up to 7 more pictures to take things from. A switch writes the full prompt for you.\", \"author\": \"AusBoss\", \"layout\": \"card\", \"accent\": \"\", \"body\": \"1. Press **Run** to try it. **Your prompt** opens with a full description, so **Auto prompt** is off.\\n2. For a short prompt, switch **Auto prompt** on. A few words are enough: \\\"a young woman outside a convenience store at night, direct flash, 35mm film\\\".\\n3. Pick a shape in **Latent Size**. **Size** stays on 16 MP.\\n4. On an RTX 5090 a 16 MP picture takes about 18 seconds, the prompt writer about 2 more.\\n5. To change a picture instead, switch **Edit a picture** on, load your picture and write the change: \\\"Make it snow.\\\"\\n\\n### Switches\\n\\n- **Edit a picture** on: the prompt changes the picture you load. Off: the prompt makes a new picture at the **Latent Size** shape.\\n- **Auto prompt** on: Qwen3.5 9B, with MiniMax's own prompt-expander instructions, turns your few words into the long description the model wants. What it wrote shows in **Show Text**. For an edit it looks at your picture and every extra picture you plugged in, and writes out the change.\\n- **High VRAM | Low VRAM** picks the model file. High VRAM: the int8 file. Low VRAM: the w4a8 file. In our tests the two looked almost the same and ran at the same speed.\\n- **Fast** on: 2 steps with the pDMD LoRA instead of the model's 6. About a third less time. It frames a little tighter and small words can come out wrong, so switch it off for posters and covers.\\n- **Upscale the result** on: SeedVR2 makes the picture twice as big on each side. Use it with Size on 4 MP.\\n- **Size: 4 MP | 16 MP** sets the size of a new picture and of an edit. 16 MP is what the model is made for. 4 MP takes about 8 seconds.\\n\\nMiniMax-H3-Image is by MiniMax, under the MiniMax-H3-Image Community License. Read it before you use the pictures for work. Powered by MiniMax-H3-Image.\\n\\nNeeds ComfyUI from the master branch (MiniMax-H3-Image support was added on 2026-11-10, after 0.44.0) and AusBoss nodes 2.22.0 or newer (update first).\", \"models\": [{\"name\": \"minimax_h3_image_pruned_int8_convrot.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"21.0 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_int8_convrot.safetensors\", \"note\": \"High VRAM\"}, {\"name\": \"minimax_h3_image_pruned_w4a8.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"12.5 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_w4a8.safetensors\", \"note\": \"Low VRAM\"}, {\"name\": \"qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors\", \"dir\": \"text_encoders\", \"size\": \"15.7 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors\", \"note\": \"both\"}, {\"name\": \"minimax_h3_image_vae_fp16.safetensors\", \"dir\": \"vae\", \"size\": \"1.2 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/vae/minimax_h3_image_vae_fp16.safetensors\", \"note\": \"both\"}, {\"name\": \"qwen3.5_9b_w4a8_gptq.safetensors\", \"dir\": \"text_encoders\", \"size\": \"7.3 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3.5_9b_w4a8_gptq.safetensors\", \"note\": \"only for Auto prompt\"}, {\"name\": \"minimax_h3_pdmd_2nfe_comfyui.safetensors\", \"dir\": \"loras\", \"size\": \"1.96 GB\", \"url\": \"https://huggingface.co/akhaliq/p-h3-image-pdmd-2step/resolve/main/minimax_h3_pdmd_2nfe_comfyui.safetensors\", \"note\": \"only for Fast\"}, {\"name\": \"seedvr2_3b_int8_convrot.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"3.46 GB\", \"url\": \"https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/diffusion_models/seedvr2_3b_int8_convrot.safetensors\", \"note\": \"only for Upscale the result\"}, {\"name\": \"seedvr2_ema_vae_fp16.safetensors\", \"dir\": \"vae\", \"size\": \"501 MB\", \"url\": \"https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/vae/seedvr2_ema_vae_fp16.safetensors\", \"note\": \"only for Upscale the result: SeedVR2's own VAE\"}], \"packs\": [{\"name\": \"ComfyUI-AusBoss\", \"url\": \"https://github.com/ausboss/ComfyUI-AusBoss\", \"node\": \"AUSBOSS_NODES_ImageCropRotatePad\"}], \"links\": [{\"label\": \"YouTube · ausbossdev\", \"url\": \"https://www.youtube.com/@ausbossdev\"}, {\"label\": \"Discord · AusBoss\", \"url\": \"https://discord.gg/RwVxY9Fuaj\"}, {\"label\": \"MiniMax H3 Image files · Hugging Face\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image\"}, {\"label\": \"GitHub · ausboss\", \"url\": \"https://github.com/ausboss\"}, {\"label\": \"Civitai · AusBoss\", \"url\": \"https://civitai.com/user/AusBoss\"}, {\"label\": \"X · Zanzibased\", \"url\": \"https://x.com/Zanzibased\"}]}"
43 ],
44 "widgets_values_named": {
45 "note": "{\"v\": 1, \"title\": \"MiniMax H3 Image · Guide + models\", \"subtitle\": \"Make a 16 megapixel picture from a few words, or change a picture you load, with MiniMax-H3-Image. Give it up to 7 more pictures to take things from. A switch writes the full prompt for you.\", \"author\": \"AusBoss\", \"layout\": \"card\", \"accent\": \"\", \"body\": \"1. Press **Run** to try it. **Your prompt** opens with a full description, so **Auto prompt** is off.\\n2. For a short prompt, switch **Auto prompt** on. A few words are enough: \\\"a young woman outside a convenience store at night, direct flash, 35mm film\\\".\\n3. Pick a shape in **Latent Size**. **Size** stays on 16 MP.\\n4. On an RTX 5090 a 16 MP picture takes about 18 seconds, the prompt writer about 2 more.\\n5. To change a picture instead, switch **Edit a picture** on, load your picture and write the change: \\\"Make it snow.\\\"\\n\\n### Switches\\n\\n- **Edit a picture** on: the prompt changes the picture you load. Off: the prompt makes a new picture at the **Latent Size** shape.\\n- **Auto prompt** on: Qwen3.5 9B, with MiniMax's own prompt-expander instructions, turns your few words into the long description the model wants. What it wrote shows in **Show Text**. For an edit it looks at your picture and every extra picture you plugged in, and writes out the change.\\n- **High VRAM | Low VRAM** picks the model file. High VRAM: the int8 file. Low VRAM: the w4a8 file. In our tests the two looked almost the same and ran at the same speed.\\n- **Fast** on: 2 steps with the pDMD LoRA instead of the model's 6. About a third less time. It frames a little tighter and small words can come out wrong, so switch it off for posters and covers.\\n- **Upscale the result** on: SeedVR2 makes the picture twice as big on each side. Use it with Size on 4 MP.\\n- **Size: 4 MP | 16 MP** sets the size of a new picture and of an edit. 16 MP is what the model is made for. 4 MP takes about 8 seconds.\\n\\nMiniMax-H3-Image is by MiniMax, under the MiniMax-H3-Image Community License. Read it before you use the pictures for work. Powered by MiniMax-H3-Image.\\n\\nNeeds ComfyUI from the master branch (MiniMax-H3-Image support was added on 2026-11-10, after 0.44.0) and AusBoss nodes 2.22.0 or newer (update first).\", \"models\": [{\"name\": \"minimax_h3_image_pruned_int8_convrot.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"21.0 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_int8_convrot.safetensors\", \"note\": \"High VRAM\"}, {\"name\": \"minimax_h3_image_pruned_w4a8.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"12.5 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_w4a8.safetensors\", \"note\": \"Low VRAM\"}, {\"name\": \"qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors\", \"dir\": \"text_encoders\", \"size\": \"15.7 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors\", \"note\": \"both\"}, {\"name\": \"minimax_h3_image_vae_fp16.safetensors\", \"dir\": \"vae\", \"size\": \"1.2 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/vae/minimax_h3_image_vae_fp16.safetensors\", \"note\": \"both\"}, {\"name\": \"qwen3.5_9b_w4a8_gptq.safetensors\", \"dir\": \"text_encoders\", \"size\": \"7.3 GB\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3.5_9b_w4a8_gptq.safetensors\", \"note\": \"only for Auto prompt\"}, {\"name\": \"minimax_h3_pdmd_2nfe_comfyui.safetensors\", \"dir\": \"loras\", \"size\": \"1.96 GB\", \"url\": \"https://huggingface.co/akhaliq/p-h3-image-pdmd-2step/resolve/main/minimax_h3_pdmd_2nfe_comfyui.safetensors\", \"note\": \"only for Fast\"}, {\"name\": \"seedvr2_3b_int8_convrot.safetensors\", \"dir\": \"diffusion_models\", \"size\": \"3.46 GB\", \"url\": \"https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/diffusion_models/seedvr2_3b_int8_convrot.safetensors\", \"note\": \"only for Upscale the result\"}, {\"name\": \"seedvr2_ema_vae_fp16.safetensors\", \"dir\": \"vae\", \"size\": \"501 MB\", \"url\": \"https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/vae/seedvr2_ema_vae_fp16.safetensors\", \"note\": \"only for Upscale the result: SeedVR2's own VAE\"}], \"packs\": [{\"name\": \"ComfyUI-AusBoss\", \"url\": \"https://github.com/ausboss/ComfyUI-AusBoss\", \"node\": \"AUSBOSS_NODES_ImageCropRotatePad\"}], \"links\": [{\"label\": \"YouTube · ausbossdev\", \"url\": \"https://www.youtube.com/@ausbossdev\"}, {\"label\": \"Discord · AusBoss\", \"url\": \"https://discord.gg/RwVxY9Fuaj\"}, {\"label\": \"MiniMax H3 Image files · Hugging Face\", \"url\": \"https://huggingface.co/Comfy-Org/MiniMax-H3-Image\"}, {\"label\": \"GitHub · ausboss\", \"url\": \"https://github.com/ausboss\"}, {\"label\": \"Civitai · AusBoss\", \"url\": \"https://civitai.com/user/AusBoss\"}, {\"label\": \"X · Zanzibased\", \"url\": \"https://x.com/Zanzibased\"}]}"
46 },
47 "color": "#14424d",
48 "bgcolor": "#161f21"
49 },
50 {
51 "id": 3,
52 "type": "AUSBOSS_NODES_Text",
53 "pos": [
54 500,
55 80
56 ],
57 "size": [
58 424,
59 300
60 ],
61 "flags": [],
62 "order": 2,
63 "mode": 0,
64 "inputs": [
65 {
66 "localized_name": "text",
67 "name": "text",
68 "type": "STRING",
69 "widget": {
70 "name": "text"
71 },
72 "link": null
73 }
74 ],
75 "outputs": [
76 {
77 "localized_name": "text",
78 "name": "text",
79 "type": "STRING",
80 "links": [
81 1000
82 ]
83 }
84 ],
85 "title": "Text - Your prompt",
86 "properties": {
87 "Node name for S&R": "AUSBOSS_NODES_Text",
88 "cnr_id": "ausboss-nodes",
89 "ver": "2.15.0"
90 },
91 "widgets_values": [
92 "35mm film photograph at night, direct on-camera flash. A young woman in her twenties with a blunt black bob and straight bangs stands outside a convenience store window, holding a can of iced coffee. She wears a cropped gray knit sweater, a short pleated black skirt and big silver headphones around her neck. Behind her a red neon sign in the window reads \"OPEN 24 HOURS\", and a paper sign taped to the glass says \"ICED COFFEE 2 FOR $3\" in black marker. Cool, distant expression, lips closed, looking into the lens. Teal-green and amber light, deep shadows, fine film grain."
93 ],
94 "widgets_values_named": {
95 "text": "35mm film photograph at night, direct on-camera flash. A young woman in her twenties with a blunt black bob and straight bangs stands outside a convenience store window, holding a can of iced coffee. She wears a cropped gray knit sweater, a short pleated black skirt and big silver headphones around her neck. Behind her a red neon sign in the window reads \"OPEN 24 HOURS\", and a paper sign taped to the glass says \"ICED COFFEE 2 FOR $3\" in black marker. Cool, distant expression, lips closed, looking into the lens. Teal-green and amber light, deep shadows, fine film grain."
96 },
97 "color": "#14424d",
98 "bgcolor": "#161f21"
99 },
100 {
101 "id": 42,
102 "type": "AUSBOSS_NODES_Callout",
103 "pos": [
104 500,
105 450
106 ],
107 "size": [
108 424,
109 194
110 ],
111 "flags": {
112 "pinned": true
113 },
114 "order": 3,
115 "mode": 0,
116 "inputs": [
117 {
118 "localized_name": "text",
119 "name": "text",
120 "type": "STRING",
121 "widget": {
122 "name": "text"
123 },
124 "link": null
125 }
126 ],
127 "outputs": [],
128 "title": "Callout",
129 "properties": {
130 "Node name for S&R": "AUSBOSS_NODES_Callout",
131 "cnr_id": "ausboss-nodes"
132 },
133 "widgets_values": [
134 "⬆️#3 Write your prompt. With **Auto prompt** on, a few words are enough.\n###### With **Edit a picture** on, write the change instead: \"Make it snow.\"\n\n⬇️#6 For an edit, load your **picture** here.\n###### Drag the blue handles to crop it."
135 ],
136 "widgets_values_named": {
137 "text": "⬆️#3 Write your prompt. With **Auto prompt** on, a few words are enough.\n###### With **Edit a picture** on, write the change instead: \"Make it snow.\"\n\n⬇️#6 For an edit, load your **picture** here.\n###### Drag the blue handles to crop it."
138 },
139 "color": "#07090a",
140 "bgcolor": "#07090a"
141 },
142 {
143 "id": 6,
144 "type": "AUSBOSS_NODES_ImageCropRotatePad",
145 "pos": [
146 500,
147 748
148 ],
149 "size": [
150 424,
151 1100
152 ],
153 "flags": [],
154 "order": 4,
155 "mode": 4,
156 "inputs": [
157 {
158 "localized_name": "image",
159 "name": "image",
160 "type": "COMBO",
161 "widget": {
162 "name": "image"
163 },
164 "link": null
165 },
166 {
167 "localized_name": "rotation_degrees",
168 "name": "rotation_degrees",
169 "type": "FLOAT",
170 "widget": {
171 "name": "rotation_degrees"
172 },
173 "link": null
174 },
175 {
176 "localized_name": "crop_aspect_ratio",
177 "name": "crop_aspect_ratio",
178 "type": "COMBO",
179 "widget": {
180 "name": "crop_aspect_ratio"
181 },
182 "link": null
183 },
184 {
185 "localized_name": "crop_x",
186 "name": "crop_x",
187 "type": "INT",
188 "widget": {
189 "name": "crop_x"
190 },
191 "link": null
192 },
193 {
194 "localized_name": "crop_y",
195 "name": "crop_y",
196 "type": "INT",
197 "widget": {
198 "name": "crop_y"
199 },
200 "link": null
201 },
202 {
203 "localized_name": "crop_width",
204 "name": "crop_width",
205 "type": "INT",
206 "widget": {
207 "name": "crop_width"
208 },
209 "link": null
210 },
211 {
212 "localized_name": "crop_height",
213 "name": "crop_height",
214 "type": "INT",
215 "widget": {
216 "name": "crop_height"
217 },
218 "link": null
219 },
220 {
221 "localized_name": "pad_left",
222 "name": "pad_left",
223 "type": "INT",
224 "widget": {
225 "name": "pad_left"
226 },
227 "link": null
228 },
229 {
230 "localized_name": "pad_top",
231 "name": "pad_top",
232 "type": "INT",
233 "widget": {
234 "name": "pad_top"
235 },
236 "link": null
237 },
238 {
239 "localized_name": "pad_right",
240 "name": "pad_right",
241 "type": "INT",
242 "widget": {
243 "name": "pad_right"
244 },
245 "link": null
246 },
247 {
248 "localized_name": "pad_bottom",
249 "name": "pad_bottom",
250 "type": "INT",
251 "widget": {
252 "name": "pad_bottom"
253 },
254 "link": null
255 },
256 {
257 "localized_name": "feather",
258 "name": "feather",
259 "type": "INT",
260 "widget": {
261 "name": "feather"
262 },
263 "link": null
264 },
265 {
266 "localized_name": "canvas_multiple",
267 "name": "canvas_multiple",
268 "type": "INT",
269 "widget": {
270 "name": "canvas_multiple"
271 },
272 "link": null
273 },
274 {
275 "localized_name": "fill_color",
276 "name": "fill_color",
277 "type": "STRING",
278 "widget": {
279 "name": "fill_color"
280 },
281 "link": null
282 },
283 {
284 "localized_name": "resize_to_megapixels",
285 "name": "resize_to_megapixels",
286 "shape": 7,
287 "type": "BOOLEAN",
288 "widget": {
289 "name": "resize_to_megapixels"
290 },
291 "link": null
292 },
293 {
294 "localized_name": "megapixels",
295 "name": "megapixels",
296 "shape": 7,
297 "type": "FLOAT",
298 "widget": {
299 "name": "megapixels"
300 },
301 "link": null
302 },
303 {
304 "localized_name": "resize_method",
305 "name": "resize_method",
306 "shape": 7,
307 "type": "COMBO",
308 "widget": {
309 "name": "resize_method"
310 },
311 "link": null
312 },
313 {
314 "localized_name": "resolution_steps",
315 "name": "resolution_steps",
316 "shape": 7,
317 "type": "INT",
318 "widget": {
319 "name": "resolution_steps"
320 },
321 "link": null
322 },
323 {
324 "localized_name": "stitch_blend",
325 "name": "stitch_blend",
326 "shape": 7,
327 "type": "INT",
328 "widget": {
329 "name": "stitch_blend"
330 },
331 "link": null
332 },
333 {
334 "localized_name": "stitch_grow",
335 "name": "stitch_grow",
336 "shape": 7,
337 "type": "INT",
338 "widget": {
339 "name": "stitch_grow"
340 },
341 "link": null
342 },
343 {
344 "localized_name": "painted_area",
345 "name": "painted_area",
346 "shape": 7,
347 "type": "COMBO",
348 "widget": {
349 "name": "painted_area"
350 },
351 "link": null
352 },
353 {
354 "localized_name": "upload",
355 "name": "upload",
356 "type": "IMAGEUPLOAD",
357 "widget": {
358 "name": "upload"
359 },
360 "link": null
361 }
362 ],
363 "outputs": [
364 {
365 "localized_name": "image",
366 "name": "image",
367 "type": "IMAGE",
368 "links": [
369 1001,
370 1015
371 ]
372 },
373 {
374 "localized_name": "mask",
375 "name": "mask",
376 "type": "MASK",
377 "links": []
378 },
379 {
380 "localized_name": "stitcher",
381 "name": "stitcher",
382 "type": "AUSBOSS_STITCHER",
383 "links": []
384 },
385 {
386 "localized_name": "original",
387 "name": "original",
388 "type": "IMAGE",
389 "links": []
390 },
391 {
392 "localized_name": "width",
393 "name": "width",
394 "type": "INT",
395 "links": []
396 },
397 {
398 "localized_name": "height",
399 "name": "height",
400 "type": "INT",
401 "links": []
402 },
403 {
404 "localized_name": "prompt_image",
405 "name": "prompt_image",
406 "type": "IMAGE",
407 "links": []
408 }
409 ],
410 "title": "Image Crop + Rotate + Pad - Your picture",
411 "properties": {
412 "Node name for S&R": "AUSBOSS_NODES_ImageCropRotatePad",
413 "cnr_id": "ausboss-nodes",
414 "ver": "2.15.0"
415 },
416 "widgets_values": {
417 "image": "",
418 "rotation_degrees": 0,
419 "crop_aspect_ratio": "free",
420 "crop_x": 0,
421 "crop_y": 0,
422 "crop_width": 0,
423 "crop_height": 0,
424 "pad_left": 0,
425 "pad_top": 0,
426 "pad_right": 0,
427 "pad_bottom": 0,
428 "feather": 24,
429 "canvas_multiple": 1,
430 "fill_color": "#808080",
431 "resize_to_megapixels": true,
432 "megapixels": 16,
433 "resize_method": "lanczos",
434 "resolution_steps": 32,
435 "stitch_blend": 32,
436 "stitch_grow": 0,
437 "painted_area": "fill"
438 },
439 "widgets_values_named": {
440 "image": "",
441 "rotation_degrees": 0,
442 "crop_aspect_ratio": "free",
443 "crop_x": 0,
444 "crop_y": 0,
445 "crop_width": 0,
446 "crop_height": 0,
447 "pad_left": 0,
448 "pad_top": 0,
449 "pad_right": 0,
450 "pad_bottom": 0,
451 "feather": 24,
452 "canvas_multiple": 1,
453 "fill_color": "#808080",
454 "resize_to_megapixels": true,
455 "megapixels": 16,
456 "resize_method": "lanczos",
457 "resolution_steps": 32,
458 "stitch_blend": 32,
459 "stitch_grow": 0,
460 "painted_area": "fill",
461 "upload": "image"
462 },
463 "color": "#14424d",
464 "bgcolor": "#161f21"
465 },
466 {
467 "id": 43,
468 "type": "AUSBOSS_NODES_Callout",
469 "pos": [
470 964,
471 80
472 ],
473 "size": [
474 424,
475 290
476 ],
477 "flags": {
478 "pinned": true
479 },
480 "order": 5,
481 "mode": 0,
482 "inputs": [
483 {
484 "localized_name": "text",
485 "name": "text",
486 "type": "STRING",
487 "widget": {
488 "name": "text"
489 },
490 "link": null
491 }
492 ],
493 "outputs": [],
494 "title": "Callout",
495 "properties": {
496 "Node name for S&R": "AUSBOSS_NODES_Callout",
497 "cnr_id": "ausboss-nodes"
498 },
499 "widgets_values": [
500 "⬇️#2 **Edit a picture** on: the prompt changes the picture you load. Off: it makes a new picture.\n\n⬇️#2 **Auto prompt** on: a writer turns your few words into the long description MiniMax-H3-Image wants. For an edit it looks at your pictures and writes out the change.\n\n⬇️#2 **Fast** on: 2 steps with the pDMD LoRA instead of 6. Quicker, a little less exact."
501 ],
502 "widgets_values_named": {
503 "text": "⬇️#2 **Edit a picture** on: the prompt changes the picture you load. Off: it makes a new picture.\n\n⬇️#2 **Auto prompt** on: a writer turns your few words into the long description MiniMax-H3-Image wants. For an edit it looks at your pictures and writes out the change.\n\n⬇️#2 **Fast** on: 2 steps with the pDMD LoRA instead of 6. Quicker, a little less exact."
504 },
505 "color": "#07090a",
506 "bgcolor": "#07090a"
507 },
508 {
509 "id": 2,
510 "type": "AUSBOSS_NODES_WorkflowSwitches",
511 "pos": [
512 964,
513 440
514 ],
515 "size": [
516 424,
517 128
518 ],
519 "flags": [],
520 "order": 6,
521 "mode": 0,
522 "inputs": [],
523 "outputs": [],
524 "title": "Workflow Switches - What to run",
525 "properties": {
526 "Node name for S&R": "AUSBOSS_NODES_WorkflowSwitches",
527 "ausboss_workflow_switches": {
528 "off": "bypass",
529 "groups": "none",
530 "match": "",
531 "order": "canvas",
532 "exclusive": false,
533 "switches": [
534 {
535 "id": 1,
536 "title": "Edit a picture",
537 "nodes": [
538 6
539 ],
540 "groups": [],
541 "nodes_off": [
542 5,
543 308,
544 309
545 ],
546 "on": {
547 "39": {
548 "size from your picture": true
549 },
550 "41": {
551 "write an edit": true
552 }
553 },
554 "off": {
555 "39": {
556 "size from your picture": false
557 },
558 "41": {
559 "write an edit": false
560 }
561 }
562 },
563 {
564 "id": 3,
565 "title": "Auto prompt",
566 "nodes": [
567 70
568 ],
569 "groups": [],
570 "on": {
571 "41": {
572 "auto prompt": true
573 }
574 },
575 "off": {
576 "41": {
577 "auto prompt": false
578 }
579 }
580 },
581 {
582 "id": 2,
583 "title": "Upscale the result",
584 "nodes": [
585 105,
586 106,
587 107,
588 12
589 ],
590 "groups": [],
591 "on": [],
592 "off": []
593 },
594 {
595 "id": 4,
596 "title": "Fast: pDMD, 2 steps",
597 "nodes": [],
598 "groups": [],
599 "on": {
600 "11": {
601 "loras": "[{\"name\":\"minimax_h3_pdmd_2nfe_comfyui.safetensors\",\"strength\":1,\"enabled\":true,\"triggers\":\"\"}]"
602 },
603 "39": {
604 "fast": true
605 }
606 },
607 "off": {
608 "11": {
609 "loras": "[{\"name\":\"minimax_h3_pdmd_2nfe_comfyui.safetensors\",\"strength\":1,\"enabled\":false,\"triggers\":\"\"}]"
610 },
611 "39": {
612 "fast": false
613 }
614 }
615 }
616 ]
617 },
618 "ausboss_show_edit": false,
619 "cnr_id": "ausboss-nodes",
620 "ver": "2.15.0"
621 },
622 "widgets_values": [],
623 "widgets_values_named": [],
624 "color": "#14424d",
625 "bgcolor": "#161f21"
626 },
627 {
628 "id": 301,
629 "type": "AUSBOSS_NODES_Callout",
630 "pos": [
631 964,
632 620
633 ],
634 "size": [
635 424,
636 122
637 ],
638 "flags": {
639 "pinned": true
640 },
641 "order": 7,
642 "mode": 0,
643 "inputs": [
644 {
645 "localized_name": "text",
646 "name": "text",
647 "type": "STRING",
648 "widget": {
649 "name": "text"
650 },
651 "link": null
652 }
653 ],
654 "outputs": [],
655 "title": "Callout",
656 "properties": {
657 "Node name for S&R": "AUSBOSS_NODES_Callout",
658 "cnr_id": "ausboss-nodes"
659 },
660 "widgets_values": [
661 "⬇️#302 **VRAM**: pick the model file for your card.\n###### High VRAM: the int8 file. Low VRAM: the smaller w4a8 file. Get them from the note ⬅️#1"
662 ],
663 "widgets_values_named": {
664 "text": "⬇️#302 **VRAM**: pick the model file for your card.\n###### High VRAM: the int8 file. Low VRAM: the smaller w4a8 file. Get them from the note ⬅️#1"
665 },
666 "color": "#07090a",
667 "bgcolor": "#07090a"
668 },
669 {
670 "id": 302,
671 "type": "AUSBOSS_NODES_WorkflowSwitches",
672 "pos": [
673 964,
674 790
675 ],
676 "size": [
677 424,
678 66
679 ],
680 "flags": [],
681 "order": 8,
682 "mode": 0,
683 "inputs": [],
684 "outputs": [],
685 "title": "Workflow Switches - VRAM",
686 "properties": {
687 "Node name for S&R": "AUSBOSS_NODES_WorkflowSwitches",
688 "ausboss_workflow_switches": {
689 "off": "bypass",
690 "groups": "none",
691 "match": "",
692 "order": "canvas",
693 "exclusive": true,
694 "show": "pill",
695 "switches": [
696 {
697 "id": 1,
698 "title": "High VRAM: 32 GB+",
699 "nodes": [],
700 "groups": [],
701 "on": {
702 "10": {
703 "unet_name": "minimax_h3_image_pruned_int8_convrot.safetensors"
704 }
705 },
706 "off": {
707 "10": {
708 "unet_name": "minimax_h3_image_pruned_w4a8.safetensors"
709 }
710 }
711 },
712 {
713 "id": 2,
714 "title": "Low VRAM: 24 GB",
715 "nodes": [],
716 "groups": [],
717 "on": {
718 "10": {
719 "unet_name": "minimax_h3_image_pruned_w4a8.safetensors"
720 }
721 },
722 "off": {
723 "10": {
724 "unet_name": "minimax_h3_image_pruned_int8_convrot.safetensors"
725 }
726 }
727 }
728 ]
729 },
730 "ausboss_show_edit": false,
731 "cnr_id": "ausboss-nodes",
732 "ver": "2.15.0"
733 },
734 "widgets_values": [],
735 "widgets_values_named": [],
736 "color": "#14424d",
737 "bgcolor": "#161f21"
738 },
739 {
740 "id": 4,
741 "type": "AUSBOSS_NODES_Seed",
742 "pos": [
743 964,
744 1260
745 ],
746 "size": [
747 424,
748 216
749 ],
750 "flags": [],
751 "order": 9,
752 "mode": 0,
753 "inputs": [
754 {
755 "localized_name": "seed",
756 "name": "seed",
757 "type": "INT",
758 "widget": {
759 "name": "seed"
760 },
761 "link": null
762 }
763 ],
764 "outputs": [
765 {
766 "localized_name": "seed",
767 "name": "seed",
768 "type": "INT",
769 "links": [
770 1014,
771 1023
772 ]
773 }
774 ],
775 "title": "Seed",
776 "properties": {
777 "Node name for S&R": "AUSBOSS_NODES_Seed",
778 "cnr_id": "ausboss-nodes",
779 "ver": "2.15.0"
780 },
781 "widgets_values": [
782 20261009,
783 "randomize"
784 ],
785 "widgets_values_named": {
786 "seed": 20261009,
787 "control_after_generate": "randomize"
788 },
789 "color": "#14424d",
790 "bgcolor": "#161f21"
791 },
792 {
793 "id": 5,
794 "type": "AUSBOSS_NODES_Resolution",
795 "pos": [
796 964,
797 1570
798 ],
799 "size": [
800 424,
801 677
802 ],
803 "flags": [],
804 "order": 10,
805 "mode": 0,
806 "inputs": [
807 {
808 "localized_name": "width",
809 "name": "width",
810 "type": "INT",
811 "link": null
812 },
813 {
814 "localized_name": "height",
815 "name": "height",
816 "type": "INT",
817 "link": null
818 },
819 {
820 "localized_name": "latent",
821 "name": "latent",
822 "shape": 7,
823 "type": "COMBO",
824 "link": null
825 },
826 {
827 "localized_name": "batch_size",
828 "name": "batch_size",
829 "shape": 7,
830 "type": "INT",
831 "link": null
832 }
833 ],
834 "outputs": [
835 {
836 "localized_name": "width",
837 "name": "width",
838 "type": "INT",
839 "links": [
840 1125,
841 1127
842 ]
843 },
844 {
845 "localized_name": "height",
846 "name": "height",
847 "type": "INT",
848 "links": [
849 1126,
850 1128
851 ]
852 },
853 {
854 "localized_name": "latent",
855 "name": "latent",
856 "type": "LATENT",
857 "links": []
858 }
859 ],
860 "title": "Latent Size - Shape of a new picture",
861 "properties": {
862 "Node name for S&R": "AUSBOSS_NODES_Resolution",
863 "ausbossResolutionPortrait": true,
864 "ausbossResolutionSnap": 32,
865 "cnr_id": "ausboss-nodes",
866 "ver": "2.15.0"
867 },
868 "widgets_values": [
869 3360,
870 5024,
871 "16ch",
872 1
873 ],
874 "widgets_values_named": {
875 "width": 3360,
876 "height": 5024,
877 "latent": "16ch",
878 "batch_size": 1
879 },
880 "color": "#14424d",
881 "bgcolor": "#161f21"
882 },
883 {
884 "id": 44,
885 "type": "AUSBOSS_NODES_CalloutTile",
886 "pos": [
887 1468,
888 80
889 ],
890 "size": [
891 1220,
892 110
893 ],
894 "flags": {
895 "pinned": true
896 },
897 "order": 7,
898 "mode": 0,
899 "inputs": [
900 {
901 "localized_name": "text",
902 "name": "text",
903 "type": "STRING",
904 "widget": {
905 "name": "text"
906 },
907 "link": null
908 }
909 ],
910 "outputs": [],
911 "title": "Callout",
912 "properties": {
913 "Node name for S&R": "AUSBOSS_NODES_CalloutTile",
914 "cnr_id": "ausboss-nodes",
915 "ausbossCalloutTile": "title"
916 },
917 "widgets_values": [
918 "Two boxes do the work: one writes the prompt, one makes or changes the picture.\n###### Double-click one to see inside."
919 ],
920 "widgets_values_named": {
921 "text": "Two boxes do the work: one writes the prompt, one makes or changes the picture.\n###### Double-click one to see inside."
922 },
923 "color": "#07090a",
924 "bgcolor": "#07090a"
925 },
926 {
927 "id": 303,
928 "type": "AUSBOSS_NODES_Callout",
929 "pos": [
930 1468,
931 308
932 ],
933 "size": [
934 300,
935 250
936 ],
937 "flags": {
938 "pinned": true
939 },
940 "order": 12,
941 "mode": 0,
942 "inputs": [
943 {
944 "localized_name": "text",
945 "name": "text",
946 "type": "STRING",
947 "widget": {
948 "name": "text"
949 },
950 "link": null
951 }
952 ],
953 "outputs": [],
954 "title": "Callout",
955 "properties": {
956 "Node name for S&R": "AUSBOSS_NODES_Callout",
957 "cnr_id": "ausboss-nodes"
958 },
959 "widgets_values": [
960 "➡️#41 **More pictures?** Plug any image loader into a free **picture** socket.\n###### Go in order from picture 2, up to picture 8. Say \"the hat from picture 2\" in your prompt. They work with **Edit a picture** on."
961 ],
962 "widgets_values_named": {
963 "text": "➡️#41 **More pictures?** Plug any image loader into a free **picture** socket.\n###### Go in order from picture 2, up to picture 8. Say \"the hat from picture 2\" in your prompt. They work with **Edit a picture** on."
964 },
965 "color": "#07090a",
966 "bgcolor": "#07090a"
967 },
968 {
969 "id": 41,
970 "type": "763be216-d720-52d4-b604-469834ebd74c",
971 "pos": [
972 1808,
973 308
974 ],
975 "size": [
976 300,
977 300
978 ],
979 "flags": [],
980 "order": 27,
981 "mode": 0,
982 "inputs": [
983 {
984 "localized_name": "your prompt",
985 "name": "your prompt",
986 "shape": 7,
987 "type": "*",
988 "link": 1000
989 },
990 {
991 "localized_name": "picture 2",
992 "name": "picture 2",
993 "shape": 7,
994 "type": "IMAGE",
995 "link": null
996 },
997 {
998 "localized_name": "picture 3",
999 "name": "picture 3",
1000 "shape": 7,
1001 "type": "IMAGE",
1002 "link": null
1003 },
1004 {
1005 "localized_name": "picture 4",
1006 "name": "picture 4",
1007 "shape": 7,
1008 "type": "IMAGE",
1009 "link": null
1010 },
1011 {
1012 "localized_name": "picture 5",
1013 "name": "picture 5",
1014 "shape": 7,
1015 "type": "IMAGE",
1016 "link": null
1017 },
1018 {
1019 "localized_name": "picture 6",
1020 "name": "picture 6",
1021 "shape": 7,
1022 "type": "IMAGE",
1023 "link": null
1024 },
1025 {
1026 "localized_name": "picture 7",
1027 "name": "picture 7",
1028 "shape": 7,
1029 "type": "IMAGE",
1030 "link": null
1031 },
1032 {
1033 "localized_name": "picture 8",
1034 "name": "picture 8",
1035 "shape": 7,
1036 "type": "IMAGE",
1037 "link": null
1038 },
1039 {
1040 "localized_name": "your picture",
1041 "name": "your picture",
1042 "shape": 7,
1043 "type": "IMAGE",
1044 "link": 1001
1045 },
1046 {
1047 "localized_name": "prompt writer",
1048 "name": "prompt writer",
1049 "type": "CLIP",
1050 "link": 1002
1051 },
1052 {
1053 "localized_name": "width",
1054 "name": "width",
1055 "shape": 7,
1056 "type": "INT",
1057 "link": 1129
1058 },
1059 {
1060 "localized_name": "height",
1061 "name": "height",
1062 "shape": 7,
1063 "type": "INT",
1064 "link": 1130
1065 },
1066 {
1067 "localized_name": "write an edit",
1068 "name": "write an edit",
1069 "type": "BOOLEAN",
1070 "widget": {
1071 "name": "write an edit"
1072 },
1073 "link": null
1074 },
1075 {
1076 "localized_name": "auto prompt",
1077 "name": "auto prompt",
1078 "type": "BOOLEAN",
1079 "widget": {
1080 "name": "auto prompt"
1081 },
1082 "link": null
1083 }
1084 ],
1085 "outputs": [
1086 {
1087 "localized_name": "prompt",
1088 "name": "prompt",
1089 "type": "STRING",
1090 "links": [
1091 1005,
1092 1006
1093 ]
1094 },
1095 {
1096 "localized_name": "picture 2",
1097 "name": "picture 2",
1098 "type": "IMAGE",
1099 "links": [
1100 1007
1101 ]
1102 },
1103 {
1104 "localized_name": "picture 3",
1105 "name": "picture 3",
1106 "type": "IMAGE",
1107 "links": [
1108 1008
1109 ]
1110 },
1111 {
1112 "localized_name": "picture 4",
1113 "name": "picture 4",
1114 "type": "IMAGE",
1115 "links": [
1116 1009
1117 ]
1118 },
1119 {
1120 "localized_name": "picture 5",
1121 "name": "picture 5",
1122 "type": "IMAGE",
1123 "links": [
1124 1010
1125 ]
1126 },
1127 {
1128 "localized_name": "picture 6",
1129 "name": "picture 6",
1130 "type": "IMAGE",
1131 "links": [
1132 1011
1133 ]
1134 },
1135 {
1136 "localized_name": "picture 7",
1137 "name": "picture 7",
1138 "type": "IMAGE",
1139 "links": [
1140 1012
1141 ]
1142 },
1143 {
1144 "localized_name": "picture 8",
1145 "name": "picture 8",
1146 "type": "IMAGE",
1147 "links": [
1148 1013
1149 ]
1150 }
1151 ],
1152 "properties": {
1153 "Node name for S&R": "763be216-d720-52d4-b604-469834ebd74c",
1154 "previewExposures": []
1155 },
1156 "widgets_values": [
1157 false,
1158 false
1159 ],
1160 "widgets_values_named": {
1161 "write an edit": false,
1162 "auto prompt": false
1163 },
1164 "color": "#14424d",
1165 "bgcolor": "#161f21"
1166 },
1167 {
1168 "id": 15,
1169 "type": "AUSBOSS_NODES_ShowText",
1170 "pos": [
1171 1468,
1172 628
1173 ],
1174 "size": [
1175 300,
1176 430
1177 ],
1178 "flags": [],
1179 "order": 28,
1180 "mode": 0,
1181 "inputs": [
1182 {
1183 "localized_name": "text",
1184 "name": "text",
1185 "type": "*",
1186 "link": 1005
1187 }
1188 ],
1189 "outputs": [
1190 {
1191 "localized_name": "text",
1192 "name": "text",
1193 "type": "STRING",
1194 "links": []
1195 }
1196 ],
1197 "title": "Show Text - Prompt MiniMax-H3-Image sees",
1198 "properties": {
1199 "Node name for S&R": "AUSBOSS_NODES_ShowText",
1200 "cnr_id": "ausboss-nodes",
1201 "ver": "2.15.0"
1202 },
1203 "widgets_values": [],
1204 "widgets_values_named": [],
1205 "color": "#14424d",
1206 "bgcolor": "#161f21"
1207 },
1208 {
1209 "id": 39,
1210 "type": "cb4401eb-1636-558a-bc9a-95a42a91ff55",
1211 "pos": [
1212 1808,
1213 678
1214 ],
1215 "size": [
1216 300,
1217 380
1218 ],
1219 "flags": [],
1220 "order": 29,
1221 "mode": 0,
1222 "inputs": [
1223 {
1224 "localized_name": "prompt",
1225 "name": "prompt",
1226 "type": "*",
1227 "widget": {
1228 "name": "prompt"
1229 },
1230 "link": 1006
1231 },
1232 {
1233 "localized_name": "seed",
1234 "name": "seed",
1235 "type": "*",
1236 "widget": {
1237 "name": "seed"
1238 },
1239 "link": 1014
1240 },
1241 {
1242 "localized_name": "your picture",
1243 "name": "your picture",
1244 "shape": 7,
1245 "type": "IMAGE",
1246 "link": 1015
1247 },
1248 {
1249 "localized_name": "width",
1250 "name": "width",
1251 "shape": 7,
1252 "type": "INT",
1253 "link": 1131
1254 },
1255 {
1256 "localized_name": "height",
1257 "name": "height",
1258 "shape": 7,
1259 "type": "INT",
1260 "link": 1132
1261 },
1262 {
1263 "localized_name": "text encoder",
1264 "name": "text encoder",
1265 "type": "CLIP",
1266 "link": 1018
1267 },
1268 {
1269 "localized_name": "vae",
1270 "name": "vae",
1271 "type": "VAE",
1272 "link": 1019
1273 },
1274 {
1275 "localized_name": "diffusion model",
1276 "name": "diffusion model",
1277 "type": "MODEL",
1278 "link": 1021
1279 },
1280 {
1281 "localized_name": "picture 2",
1282 "name": "picture 2",
1283 "shape": 7,
1284 "type": "IMAGE",
1285 "link": 1007
1286 },
1287 {
1288 "localized_name": "picture 3",
1289 "name": "picture 3",
1290 "shape": 7,
1291 "type": "IMAGE",
1292 "link": 1008
1293 },
1294 {
1295 "localized_name": "picture 4",
1296 "name": "picture 4",
1297 "shape": 7,
1298 "type": "IMAGE",
1299 "link": 1009
1300 },
1301 {
1302 "localized_name": "picture 5",
1303 "name": "picture 5",
1304 "shape": 7,
1305 "type": "IMAGE",
1306 "link": 1010
1307 },
1308 {
1309 "localized_name": "picture 6",
1310 "name": "picture 6",
1311 "shape": 7,
1312 "type": "IMAGE",
1313 "link": 1011
1314 },
1315 {
1316 "localized_name": "picture 7",
1317 "name": "picture 7",
1318 "shape": 7,
1319 "type": "IMAGE",
1320 "link": 1012
1321 },
1322 {
1323 "localized_name": "picture 8",
1324 "name": "picture 8",
1325 "shape": 7,
1326 "type": "IMAGE",
1327 "link": 1013
1328 },
1329 {
1330 "localized_name": "size from your picture",
1331 "name": "size from your picture",
1332 "type": "BOOLEAN",
1333 "widget": {
1334 "name": "size from your picture"
1335 },
1336 "link": null
1337 },
1338 {
1339 "localized_name": "fast",
1340 "name": "fast",
1341 "type": "BOOLEAN",
1342 "widget": {
1343 "name": "fast"
1344 },
1345 "link": null
1346 }
1347 ],
1348 "outputs": [
1349 {
1350 "localized_name": "picture",
1351 "name": "picture",
1352 "type": "IMAGE",
1353 "links": [
1354 1022,
1355 1027,
1356 1030
1357 ]
1358 },
1359 {
1360 "localized_name": "before",
1361 "name": "before",
1362 "type": "IMAGE",
1363 "links": [
1364 1029
1365 ]
1366 }
1367 ],
1368 "properties": {
1369 "Node name for S&R": "cb4401eb-1636-558a-bc9a-95a42a91ff55",
1370 "previewExposures": []
1371 },
1372 "widgets_values": [
1373 "",
1374 0,
1375 false,
1376 false
1377 ],
1378 "widgets_values_named": {
1379 "prompt": "",
1380 "seed": 0,
1381 "size from your picture": false,
1382 "fast": false
1383 },
1384 "color": "#14424d",
1385 "bgcolor": "#161f21"
1386 },
1387 {
1388 "id": 105,
1389 "type": "5a45399a-b68e-5039-90e9-93ae815a69f1",
1390 "pos": [
1391 2148,
1392 308
1393 ],
1394 "size": [
1395 540,
1396 150
1397 ],
1398 "flags": [],
1399 "order": 30,
1400 "mode": 4,
1401 "inputs": [
1402 {
1403 "localized_name": "picture",
1404 "name": "picture",
1405 "type": "IMAGE",
1406 "link": 1022
1407 },
1408 {
1409 "localized_name": "seed",
1410 "name": "seed",
1411 "type": "INT",
1412 "widget": {
1413 "name": "seed"
1414 },
1415 "link": 1023
1416 },
1417 {
1418 "localized_name": "seedvr2 model",
1419 "name": "seedvr2 model",
1420 "type": "MODEL",
1421 "link": 1024
1422 },
1423 {
1424 "localized_name": "seedvr2 vae",
1425 "name": "seedvr2 vae",
1426 "type": "VAE",
1427 "link": 1025
1428 }
1429 ],
1430 "outputs": [
1431 {
1432 "localized_name": "picture",
1433 "name": "picture",
1434 "type": "IMAGE",
1435 "links": [
1436 1026,
1437 1028
1438 ]
1439 }
1440 ],
1441 "properties": {
1442 "previewExposures": []
1443 },
1444 "widgets_values": [
1445 20261017
1446 ],
1447 "widgets_values_named": {
1448 "seed": 20261017
1449 },
1450 "color": "#14424d",
1451 "bgcolor": "#161f21"
1452 },
1453 {
1454 "id": 8,
1455 "type": "CLIPLoader",
1456 "pos": [
1457 1468,
1458 1170
1459 ],
1460 "size": [
1461 300,
1462 106
1463 ],
1464 "flags": [],
1465 "order": 13,
1466 "mode": 0,
1467 "inputs": [
1468 {
1469 "localized_name": "clip_name",
1470 "name": "clip_name",
1471 "type": "COMBO",
1472 "widget": {
1473 "name": "clip_name"
1474 },
1475 "link": null
1476 },
1477 {
1478 "localized_name": "type",
1479 "name": "type",
1480 "type": "COMBO",
1481 "widget": {
1482 "name": "type"
1483 },
1484 "link": null
1485 },
1486 {
1487 "localized_name": "device",
1488 "name": "device",
1489 "shape": 7,
1490 "type": "COMBO",
1491 "widget": {
1492 "name": "device"
1493 },
1494 "link": null
1495 }
1496 ],
1497 "outputs": [
1498 {
1499 "localized_name": "CLIP",
1500 "name": "CLIP",
1501 "type": "CLIP",
1502 "links": [
1503 1018
1504 ]
1505 }
1506 ],
1507 "title": "Text encoder",
1508 "properties": {
1509 "Node name for S&R": "CLIPLoader",
1510 "models": [
1511 {
1512 "name": "qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
1513 "url": "https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
1514 "directory": "text_encoders"
1515 }
1516 ],
1517 "cnr_id": "comfy-core",
1518 "ver": "ede4e5fe8"
1519 },
1520 "widgets_values": [
1521 "qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
1522 "minimax",
1523 "default"
1524 ],
1525 "widgets_values_named": {
1526 "clip_name": "qwen3vl_32b_minimax_h3_nvfp4_awq.safetensors",
1527 "type": "minimax",
1528 "device": "default"
1529 },
1530 "color": "#432",
1531 "bgcolor": "#653"
1532 },
1533 {
1534 "id": 9,
1535 "type": "VAELoader",
1536 "pos": [
1537 1468,
1538 1354
1539 ],
1540 "size": [
1541 300,
1542 58
1543 ],
1544 "flags": [],
1545 "order": 14,
1546 "mode": 0,
1547 "inputs": [
1548 {
1549 "localized_name": "vae_name",
1550 "name": "vae_name",
1551 "type": "COMBO",
1552 "widget": {
1553 "name": "vae_name"
1554 },
1555 "link": null
1556 }
1557 ],
1558 "outputs": [
1559 {
1560 "localized_name": "VAE",
1561 "name": "VAE",
1562 "type": "VAE",
1563 "links": [
1564 1019
1565 ]
1566 }
1567 ],
1568 "title": "VAE",
1569 "properties": {
1570 "Node name for S&R": "VAELoader",
1571 "models": [
1572 {
1573 "name": "minimax_h3_image_vae_fp16.safetensors",
1574 "url": "https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/vae/minimax_h3_image_vae_fp16.safetensors",
1575 "directory": "vae"
1576 }
1577 ],
1578 "cnr_id": "comfy-core",
1579 "ver": "ede4e5fe8"
1580 },
1581 "widgets_values": [
1582 "minimax_h3_image_vae_fp16.safetensors"
1583 ],
1584 "widgets_values_named": {
1585 "vae_name": "minimax_h3_image_vae_fp16.safetensors"
1586 },
1587 "color": "#322",
1588 "bgcolor": "#533"
1589 },
1590 {
1591 "id": 10,
1592 "type": "UNETLoader",
1593 "pos": [
1594 1468,
1595 1490
1596 ],
1597 "size": [
1598 300,
1599 82
1600 ],
1601 "flags": [],
1602 "order": 15,
1603 "mode": 0,
1604 "inputs": [
1605 {
1606 "localized_name": "unet_name",
1607 "name": "unet_name",
1608 "type": "COMBO",
1609 "widget": {
1610 "name": "unet_name"
1611 },
1612 "link": null
1613 },
1614 {
1615 "localized_name": "weight_dtype",
1616 "name": "weight_dtype",
1617 "type": "COMBO",
1618 "widget": {
1619 "name": "weight_dtype"
1620 },
1621 "link": null
1622 }
1623 ],
1624 "outputs": [
1625 {
1626 "localized_name": "MODEL",
1627 "name": "MODEL",
1628 "type": "MODEL",
1629 "links": [
1630 1020
1631 ]
1632 }
1633 ],
1634 "title": "Diffusion model",
1635 "properties": {
1636 "Node name for S&R": "UNETLoader",
1637 "models": [
1638 {
1639 "name": "minimax_h3_image_pruned_int8_convrot.safetensors",
1640 "url": "https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/diffusion_models/minimax_h3_image_pruned_int8_convrot.safetensors",
1641 "directory": "diffusion_models"
1642 }
1643 ],
1644 "cnr_id": "comfy-core",
1645 "ver": "ede4e5fe8"
1646 },
1647 "widgets_values": [
1648 "minimax_h3_image_pruned_int8_convrot.safetensors",
1649 "default"
1650 ],
1651 "widgets_values_named": {
1652 "unet_name": "minimax_h3_image_pruned_int8_convrot.safetensors",
1653 "weight_dtype": "default"
1654 },
1655 "color": "#323",
1656 "bgcolor": "#535"
1657 },
1658 {
1659 "id": 70,
1660 "type": "CLIPLoader",
1661 "pos": [
1662 1468,
1663 1650
1664 ],
1665 "size": [
1666 300,
1667 106
1668 ],
1669 "flags": [],
1670 "order": 16,
1671 "mode": 4,
1672 "inputs": [
1673 {
1674 "localized_name": "clip_name",
1675 "name": "clip_name",
1676 "type": "COMBO",
1677 "widget": {
1678 "name": "clip_name"
1679 },
1680 "link": null
1681 },
1682 {
1683 "localized_name": "type",
1684 "name": "type",
1685 "type": "COMBO",
1686 "widget": {
1687 "name": "type"
1688 },
1689 "link": null
1690 },
1691 {
1692 "localized_name": "device",
1693 "name": "device",
1694 "shape": 7,
1695 "type": "COMBO",
1696 "widget": {
1697 "name": "device"
1698 },
1699 "link": null
1700 }
1701 ],
1702 "outputs": [
1703 {
1704 "localized_name": "CLIP",
1705 "name": "CLIP",
1706 "type": "CLIP",
1707 "links": [
1708 1002
1709 ]
1710 }
1711 ],
1712 "title": "Prompt writer - Auto prompt",
1713 "properties": {
1714 "Node name for S&R": "CLIPLoader",
1715 "models": [
1716 {
1717 "name": "qwen3.5_9b_w4a8_gptq.safetensors",
1718 "url": "https://huggingface.co/Comfy-Org/MiniMax-H3-Image/resolve/main/text_encoders/qwen3.5_9b_w4a8_gptq.safetensors",
1719 "directory": "text_encoders"
1720 }
1721 ],
1722 "cnr_id": "comfy-core",
1723 "ver": "ede4e5fe8"
1724 },
1725 "widgets_values": [
1726 "qwen3.5_9b_w4a8_gptq.safetensors",
1727 "minimax",
1728 "default"
1729 ],
1730 "widgets_values_named": {
1731 "clip_name": "qwen3.5_9b_w4a8_gptq.safetensors",
1732 "type": "minimax",
1733 "device": "default"
1734 },
1735 "color": "#432",
1736 "bgcolor": "#653"
1737 },
1738 {
1739 "id": 45,
1740 "type": "AUSBOSS_NODES_Callout",
1741 "pos": [
1742 1808,
1743 1170
1744 ],
1745 "size": [
1746 560,
1747 150
1748 ],
1749 "flags": {
1750 "pinned": true
1751 },
1752 "order": 17,
1753 "mode": 0,
1754 "inputs": [
1755 {
1756 "localized_name": "text",
1757 "name": "text",
1758 "type": "STRING",
1759 "widget": {
1760 "name": "text"
1761 },
1762 "link": null
1763 }
1764 ],
1765 "outputs": [],
1766 "title": "Callout",
1767 "properties": {
1768 "Node name for S&R": "AUSBOSS_NODES_Callout",
1769 "cnr_id": "ausboss-nodes"
1770 },
1771 "widgets_values": [
1772 "⬇️#11 The locked row is the **Fast** LoRA: the switch turns it on. **Your LoRAs** go below it. Get the files from the note ⬅️#1\n###### The prompt writer loads only with **Auto prompt** on."
1773 ],
1774 "widgets_values_named": {
1775 "text": "⬇️#11 The locked row is the **Fast** LoRA: the switch turns it on. **Your LoRAs** go below it. Get the files from the note ⬅️#1\n###### The prompt writer loads only with **Auto prompt** on."
1776 },
1777 "color": "#07090a",
1778 "bgcolor": "#07090a"
1779 },
1780 {
1781 "id": 11,
1782 "type": "AUSBOSS_NODES_LoraLoader",
1783 "pos": [
1784 1808,
1785 1390
1786 ],
1787 "size": [
1788 560,
1789 202
1790 ],
1791 "flags": [],
1792 "order": 26,
1793 "mode": 0,
1794 "inputs": [
1795 {
1796 "localized_name": "model",
1797 "name": "model",
1798 "type": "MODEL",
1799 "link": 1020
1800 },
1801 {
1802 "localized_name": "clip",
1803 "name": "clip",
1804 "shape": 7,
1805 "type": "CLIP",
1806 "link": null
1807 },
1808 {
1809 "localized_name": "loras",
1810 "name": "loras",
1811 "type": "STRING",
1812 "widget": {
1813 "name": "loras"
1814 },
1815 "link": null
1816 },
1817 {
1818 "localized_name": "trigger_separator",
1819 "name": "trigger_separator",
1820 "shape": 7,
1821 "type": "STRING",
1822 "widget": {
1823 "name": "trigger_separator"
1824 },
1825 "link": null
1826 },
1827 {
1828 "localized_name": "on_missing",
1829 "name": "on_missing",
1830 "shape": 7,
1831 "type": "COMBO",
1832 "widget": {
1833 "name": "on_missing"
1834 },
1835 "link": null
1836 }
1837 ],
1838 "outputs": [
1839 {
1840 "localized_name": "model",
1841 "name": "model",
1842 "type": "MODEL",
1843 "links": [
1844 1021
1845 ]
1846 },
1847 {
1848 "localized_name": "clip",
1849 "name": "clip",
1850 "type": "CLIP",
1851 "links": []
1852 },
1853 {
1854 "localized_name": "triggers",
1855 "name": "triggers",
1856 "type": "STRING",
1857 "links": []
1858 }
1859 ],
1860 "title": "LoRA Loader - Speed LoRA, then yours",
1861 "properties": {
1862 "Node name for S&R": "AUSBOSS_NODES_LoraLoader",
1863 "ausbossLoraLinked": true,
1864 "models": [
1865 {
1866 "name": "minimax_h3_pdmd_2nfe_comfyui.safetensors",
1867 "url": "https://huggingface.co/akhaliq/p-h3-image-pdmd-2step/resolve/main/minimax_h3_pdmd_2nfe_comfyui.safetensors",
1868 "directory": "loras"
1869 }
1870 ],
1871 "cnr_id": "ausboss-nodes",
1872 "ver": "2.15.0"
1873 },
1874 "widgets_values": [
1875 "[{\"name\":\"minimax_h3_pdmd_2nfe_comfyui.safetensors\",\"strength\":1,\"enabled\":false,\"triggers\":\"\"}]",
1876 ", ",
1877 "error"
1878 ],
1879 "widgets_values_named": {
1880 "loras": "[{\"name\":\"minimax_h3_pdmd_2nfe_comfyui.safetensors\",\"strength\":1,\"enabled\":false,\"triggers\":\"\"}]",
1881 "trigger_separator": ", ",
1882 "on_missing": "error"
1883 },
1884 "color": "#14424d",
1885 "bgcolor": "#161f21"
1886 },
1887 {
1888 "id": 106,
1889 "type": "UNETLoader",
1890 "pos": [
1891 2148,
1892 678
1893 ],
1894 "size": [
1895 540,
1896 82
1897 ],
1898 "flags": [],
1899 "order": 18,
1900 "mode": 4,
1901 "inputs": [
1902 {
1903 "localized_name": "unet_name",
1904 "name": "unet_name",
1905 "type": "COMBO",
1906 "widget": {
1907 "name": "unet_name"
1908 },
1909 "link": null
1910 },
1911 {
1912 "localized_name": "weight_dtype",
1913 "name": "weight_dtype",
1914 "type": "COMBO",
1915 "widget": {
1916 "name": "weight_dtype"
1917 },
1918 "link": null
1919 }
1920 ],
1921 "outputs": [
1922 {
1923 "localized_name": "MODEL",
1924 "name": "MODEL",
1925 "type": "MODEL",
1926 "links": [
1927 1024
1928 ]
1929 }
1930 ],
1931 "title": "SeedVR2 model",
1932 "properties": {
1933 "Node name for S&R": "UNETLoader",
1934 "cnr_id": "comfy-core",
1935 "ver": "0.41.3",
1936 "models": [
1937 {
1938 "name": "seedvr2_3b_int8_convrot.safetensors",
1939 "url": "https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/diffusion_models/seedvr2_3b_int8_convrot.safetensors",
1940 "directory": "diffusion_models"
1941 }
1942 ]
1943 },
1944 "widgets_values": [
1945 "seedvr2_3b_int8_convrot.safetensors",
1946 "default"
1947 ],
1948 "widgets_values_named": {
1949 "unet_name": "seedvr2_3b_int8_convrot.safetensors",
1950 "weight_dtype": "default"
1951 },
1952 "color": "#323",
1953 "bgcolor": "#535"
1954 },
1955 {
1956 "id": 107,
1957 "type": "VAELoader",
1958 "pos": [
1959 2148,
1960 820
1961 ],
1962 "size": [
1963 540,
1964 58
1965 ],
1966 "flags": [],
1967 "order": 19,
1968 "mode": 4,
1969 "inputs": [
1970 {
1971 "localized_name": "vae_name",
1972 "name": "vae_name",
1973 "type": "COMBO",
1974 "widget": {
1975 "name": "vae_name"
1976 },
1977 "link": null
1978 }
1979 ],
1980 "outputs": [
1981 {
1982 "localized_name": "VAE",
1983 "name": "VAE",
1984 "type": "VAE",
1985 "links": [
1986 1025
1987 ]
1988 }
1989 ],
1990 "title": "SeedVR2 VAE",
1991 "properties": {
1992 "Node name for S&R": "VAELoader",
1993 "cnr_id": "comfy-core",
1994 "ver": "0.41.3",
1995 "models": [
1996 {
1997 "name": "seedvr2_ema_vae_fp16.safetensors",
1998 "url": "https://huggingface.co/Comfy-Org/SeedVR2/resolve/main/vae/seedvr2_ema_vae_fp16.safetensors",
1999 "directory": "vae"
2000 }
2001 ]
2002 },
2003 "widgets_values": [
2004 "seedvr2_ema_vae_fp16.safetensors"
2005 ],
2006 "widgets_values_named": {
2007 "vae_name": "seedvr2_ema_vae_fp16.safetensors"
2008 },
2009 "color": "#322",
2010 "bgcolor": "#533"
2011 },
2012 {
2013 "id": 108,
2014 "type": "AUSBOSS_NODES_Callout",
2015 "pos": [
2016 2148,
2017 508
2018 ],
2019 "size": [
2020 540,
2021 122
2022 ],
2023 "flags": {
2024 "pinned": true
2025 },
2026 "order": 20,
2027 "mode": 0,
2028 "inputs": [
2029 {
2030 "localized_name": "text",
2031 "name": "text",
2032 "type": "STRING",
2033 "widget": {
2034 "name": "text"
2035 },
2036 "link": null
2037 }
2038 ],
2039 "outputs": [],
2040 "title": "Callout",
2041 "properties": {
2042 "Node name for S&R": "AUSBOSS_NODES_Callout",
2043 "cnr_id": "ausboss-nodes"
2044 },
2045 "widgets_values": [
2046 "⬆️#105 **Upscale the result** runs SeedVR2, and ⬇️#106 needs these two files. Get them from the note ⬅️#1\n###### Use it with **Size** on 4 MP: the result is 16 MP. They load only while that switch is on."
2047 ],
2048 "widgets_values_named": {
2049 "text": "⬆️#105 **Upscale the result** runs SeedVR2, and ⬇️#106 needs these two files. Get them from the note ⬅️#1\n###### Use it with **Size** on 4 MP: the result is 16 MP. They load only while that switch is on."
2050 },
2051 "color": "#07090a",
2052 "bgcolor": "#07090a"
2053 },
2054 {
2055 "id": 13,
2056 "type": "AUSBOSS_NODES_SaveImage",
2057 "pos": [
2058 2768,
2059 80
2060 ],
2061 "size": [
2062 460,
2063 816
2064 ],
2065 "flags": [],
2066 "order": 32,
2067 "mode": 0,
2068 "inputs": [
2069 {
2070 "localized_name": "images",
2071 "name": "images",
2072 "type": "IMAGE",
2073 "link": 1026
2074 },
2075 {
2076 "localized_name": "filename",
2077 "name": "filename",
2078 "shape": 7,
2079 "type": "STRING",
2080 "link": null
2081 },
2082 {
2083 "localized_name": "caption_text",
2084 "name": "caption_text",
2085 "shape": 7,
2086 "type": "STRING",
2087 "link": null
2088 },
2089 {
2090 "localized_name": "filename_prefix",
2091 "name": "filename_prefix",
2092 "type": "STRING",
2093 "widget": {
2094 "name": "filename_prefix"
2095 },
2096 "link": null
2097 },
2098 {
2099 "localized_name": "format",
2100 "name": "format",
2101 "type": "COMBO",
2102 "widget": {
2103 "name": "format"
2104 },
2105 "link": null
2106 },
2107 {
2108 "localized_name": "save_metadata",
2109 "name": "save_metadata",
2110 "type": "BOOLEAN",
2111 "widget": {
2112 "name": "save_metadata"
2113 },
2114 "link": null
2115 },
2116 {
2117 "localized_name": "exact_name",
2118 "name": "exact_name",
2119 "shape": 7,
2120 "type": "STRING",
2121 "widget": {
2122 "name": "exact_name"
2123 },
2124 "link": null
2125 },
2126 {
2127 "localized_name": "on_existing",
2128 "name": "on_existing",
2129 "shape": 7,
2130 "type": "COMBO",
2131 "widget": {
2132 "name": "on_existing"
2133 },
2134 "link": null
2135 },
2136 {
2137 "localized_name": "output_dir",
2138 "name": "output_dir",
2139 "shape": 7,
2140 "type": "STRING",
2141 "widget": {
2142 "name": "output_dir"
2143 },
2144 "link": null
2145 },
2146 {
2147 "localized_name": "caption",
2148 "name": "caption",
2149 "shape": 7,
2150 "type": "STRING",
2151 "widget": {
2152 "name": "caption"
2153 },
2154 "link": null
2155 },
2156 {
2157 "localized_name": "name_counter",
2158 "name": "name_counter",
2159 "shape": 7,
2160 "type": "BOOLEAN",
2161 "widget": {
2162 "name": "name_counter"
2163 },
2164 "link": null
2165 },
2166 {
2167 "localized_name": "name_date",
2168 "name": "name_date",
2169 "shape": 7,
2170 "type": "BOOLEAN",
2171 "widget": {
2172 "name": "name_date"
2173 },
2174 "link": null
2175 },
2176 {
2177 "localized_name": "name_time",
2178 "name": "name_time",
2179 "shape": 7,
2180 "type": "BOOLEAN",
2181 "widget": {
2182 "name": "name_time"
2183 },
2184 "link": null
2185 },
2186 {
2187 "localized_name": "name_size",
2188 "name": "name_size",
2189 "shape": 7,
2190 "type": "BOOLEAN",
2191 "widget": {
2192 "name": "name_size"
2193 },
2194 "link": null
2195 },
2196 {
2197 "localized_name": "name_batch",
2198 "name": "name_batch",
2199 "shape": 7,
2200 "type": "BOOLEAN",
2201 "widget": {
2202 "name": "name_batch"
2203 },
2204 "link": null
2205 },
2206 {
2207 "localized_name": "preview",
2208 "name": "preview",
2209 "shape": 7,
2210 "type": "BOOLEAN",
2211 "widget": {
2212 "name": "preview"
2213 },
2214 "link": null
2215 },
2216 {
2217 "localized_name": "step_preview",
2218 "name": "step_preview",
2219 "shape": 7,
2220 "type": "BOOLEAN",
2221 "widget": {
2222 "name": "step_preview"
2223 },
2224 "link": null
2225 }
2226 ],
2227 "outputs": [
2228 {
2229 "localized_name": "file_path",
2230 "name": "file_path",
2231 "type": "STRING",
2232 "links": []
2233 },
2234 {
2235 "localized_name": "images",
2236 "name": "images",
2237 "type": "IMAGE",
2238 "links": []
2239 }
2240 ],
2241 "title": "Save Image",
2242 "properties": {
2243 "Node name for S&R": "AUSBOSS_NODES_SaveImage",
2244 "cnr_id": "ausboss-nodes",
2245 "ver": "2.15.0"
2246 },
2247 "widgets_values": {
2248 "filename_prefix": "minimax-h3-image",
2249 "format": "png",
2250 "save_metadata": true,
2251 "exact_name": "",
2252 "on_existing": "overwrite",
2253 "output_dir": "",
2254 "caption": "",
2255 "name_counter": true,
2256 "name_date": false,
2257 "name_time": false,
2258 "name_size": false,
2259 "name_batch": false,
2260 "preview": false,
2261 "step_preview": false
2262 },
2263 "widgets_values_named": {
2264 "filename_prefix": "minimax-h3-image",
2265 "format": "png",
2266 "save_metadata": true,
2267 "exact_name": "",
2268 "on_existing": "overwrite",
2269 "output_dir": "",
2270 "caption": "",
2271 "name_counter": true,
2272 "name_date": false,
2273 "name_time": false,
2274 "name_size": false,
2275 "name_batch": false,
2276 "preview": false,
2277 "step_preview": false
2278 },
2279 "color": "#14424d",
2280 "bgcolor": "#161f21"
2281 },
2282 {
2283 "id": 12,
2284 "type": "AUSBOSS_NODES_Compare",
2285 "pos": [
2286 2768,
2287 966
2288 ],
2289 "size": [
2290 460,
2291 480
2292 ],
2293 "flags": [],
2294 "order": 33,
2295 "mode": 4,
2296 "inputs": [
2297 {
2298 "localized_name": "image_a",
2299 "name": "image_a",
2300 "type": "IMAGE",
2301 "link": 1027
2302 },
2303 {
2304 "localized_name": "image_b",
2305 "name": "image_b",
2306 "type": "IMAGE",
2307 "link": 1028
2308 },
2309 {
2310 "localized_name": "filename",
2311 "name": "filename",
2312 "shape": 7,
2313 "type": "STRING",
2314 "link": null
2315 }
2316 ],
2317 "outputs": [
2318 {
2319 "localized_name": "image_a",
2320 "name": "image_a",
2321 "type": "IMAGE",
2322 "links": []
2323 },
2324 {
2325 "localized_name": "image_b",
2326 "name": "image_b",
2327 "type": "IMAGE",
2328 "links": null
2329 }
2330 ],
2331 "title": "Image Compare A/B - Before and after the upscale",
2332 "properties": {
2333 "Node name for S&R": "AUSBOSS_NODES_Compare",
2334 "ausboss_compare_mode": "slide",
2335 "cnr_id": "ausboss-nodes",
2336 "ver": "2.15.0"
2337 },
2338 "widgets_values": [],
2339 "widgets_values_named": [],
2340 "color": "#14424d",
2341 "bgcolor": "#161f21"
2342 },
2343 {
2344 "id": 54,
2345 "type": "AUSBOSS_NODES_Compare",
2346 "pos": [
2347 2768,
2348 1516
2349 ],
2350 "size": [
2351 460,
2352 480
2353 ],
2354 "flags": [],
2355 "order": 31,
2356 "mode": 0,
2357 "inputs": [
2358 {
2359 "localized_name": "image_a",
2360 "name": "image_a",
2361 "type": "IMAGE",
2362 "link": 1029
2363 },
2364 {
2365 "localized_name": "image_b",
2366 "name": "image_b",
2367 "type": "IMAGE",
2368 "link": 1030
2369 },
2370 {
2371 "localized_name": "filename",
2372 "name": "filename",
2373 "shape": 7,
2374 "type": "STRING",
2375 "link": null
2376 }
2377 ],
2378 "outputs": [
2379 {
2380 "localized_name": "image_a",
2381 "name": "image_a",
2382 "type": "IMAGE",
2383 "links": []
2384 },
2385 {
2386 "localized_name": "image_b",
2387 "name": "image_b",
2388 "type": "IMAGE",
2389 "links": null
2390 }
2391 ],
2392 "title": "Image Compare A/B - Result",
2393 "properties": {
2394 "Node name for S&R": "AUSBOSS_NODES_Compare",
2395 "ausboss_compare_mode": "slide",
2396 "cnr_id": "ausboss-nodes",
2397 "ver": "2.15.0"
2398 },
2399 "widgets_values": [],
2400 "widgets_values_named": [],
2401 "color": "#14424d",
2402 "bgcolor": "#161f21"
2403 },
2404 {
2405 "id": 14,
2406 "type": "AUSBOSS_NODES_RunTimer",
2407 "pos": [
2408 2893,
2409 2066
2410 ],
2411 "size": [
2412 210,
2413 67
2414 ],
2415 "flags": {
2416 "no_title": true
2417 },
2418 "order": 21,
2419 "mode": 0,
2420 "inputs": [],
2421 "outputs": [],
2422 "title": "Run Timer",
2423 "properties": {
2424 "Node name for S&R": "AUSBOSS_NODES_RunTimer",
2425 "cnr_id": "ausboss-nodes",
2426 "ver": "2.15.0"
2427 },
2428 "color": "#07090a",
2429 "bgcolor": "#07090a"
2430 },
2431 {
2432 "id": 304,
2433 "type": "AUSBOSS_NODES_WorkflowSwitches",
2434 "pos": [
2435 964,
2436 1140
2437 ],
2438 "size": [
2439 424,
2440 66
2441 ],
2442 "flags": [],
2443 "order": 22,
2444 "mode": 0,
2445 "inputs": [],
2446 "outputs": [],
2447 "title": "Workflow Switches - Size",
2448 "properties": {
2449 "Node name for S&R": "AUSBOSS_NODES_WorkflowSwitches",
2450 "ausboss_workflow_switches": {
2451 "off": "bypass",
2452 "groups": "none",
2453 "match": "",
2454 "order": "canvas",
2455 "exclusive": true,
2456 "show": "pill",
2457 "switches": [
2458 {
2459 "id": 1,
2460 "title": "4 MP",
2461 "nodes": [],
2462 "groups": [],
2463 "on": {
2464 "6": {
2465 "resize_to_megapixels": true,
2466 "megapixels": 4
2467 },
2468 "308": {
2469 "expression": "round(a * sqrt(4194304 / (a * b)) / 32) * 32"
2470 },
2471 "309": {
2472 "expression": "round(b * sqrt(4194304 / (a * b)) / 32) * 32"
2473 }
2474 },
2475 "off": {
2476 "6": {
2477 "resize_to_megapixels": true,
2478 "megapixels": 16
2479 },
2480 "308": {
2481 "expression": "round(a * sqrt(16777216 / (a * b)) / 32) * 32"
2482 },
2483 "309": {
2484 "expression": "round(b * sqrt(16777216 / (a * b)) / 32) * 32"
2485 }
2486 }
2487 },
2488 {
2489 "id": 2,
2490 "title": "16 MP",
2491 "nodes": [],
2492 "groups": [],
2493 "on": {
2494 "6": {
2495 "resize_to_megapixels": true,
2496 "megapixels": 16
2497 },
2498 "308": {
2499 "expression": "round(a * sqrt(16777216 / (a * b)) / 32) * 32"
2500 },
2501 "309": {
2502 "expression": "round(b * sqrt(16777216 / (a * b)) / 32) * 32"
2503 }
2504 },
2505 "off": {
2506 "6": {
2507 "resize_to_megapixels": true,
2508 "megapixels": 4
2509 },
2510 "308": {
2511 "expression": "round(a * sqrt(4194304 / (a * b)) / 32) * 32"
2512 },
2513 "309": {
2514 "expression": "round(b * sqrt(4194304 / (a * b)) / 32) * 32"
2515 }
2516 }
2517 }
2518 ]
2519 },
2520 "ausboss_show_edit": false,
2521 "cnr_id": "ausboss-nodes",
2522 "ver": "2.15.0"
2523 },
2524 "widgets_values": [],
2525 "widgets_values_named": [],
2526 "color": "#14424d",
2527 "bgcolor": "#161f21"
2528 },
2529 {
2530 "id": 305,
2531 "type": "AUSBOSS_NODES_Callout",
2532 "pos": [
2533 964,
2534 910
2535 ],
2536 "size": [
2537 424,
2538 170
2539 ],
2540 "flags": {
2541 "pinned": true
2542 },
2543 "order": 23,
2544 "mode": 0,
2545 "inputs": [
2546 {
2547 "localized_name": "text",
2548 "name": "text",
2549 "type": "STRING",
2550 "widget": {
2551 "name": "text"
2552 },
2553 "link": null
2554 }
2555 ],
2556 "outputs": [],
2557 "title": "Callout",
2558 "properties": {
2559 "Node name for S&R": "AUSBOSS_NODES_Callout",
2560 "cnr_id": "ausboss-nodes"
2561 },
2562 "widgets_values": [
2563 "⬇️#304 **Size: 4 MP | 16 MP**\n###### 16 MP is what the model is made for. 4 MP is faster and uses less memory. Latent Size sets only the shape of a new picture."
2564 ],
2565 "widgets_values_named": {
2566 "text": "⬇️#304 **Size: 4 MP | 16 MP**\n###### 16 MP is what the model is made for. 4 MP is faster and uses less memory. Latent Size sets only the shape of a new picture."
2567 },
2568 "color": "#07090a",
2569 "bgcolor": "#07090a"
2570 },
2571 {
2572 "id": 308,
2573 "type": "AUSBOSS_NODES_MathExpression",
2574 "pos": [
2575 964,
2576 2290
2577 ],
2578 "size": [
2579 300,
2580 134
2581 ],
2582 "flags": {
2583 "collapsed": true
2584 },
2585 "order": 24,
2586 "mode": 0,
2587 "inputs": [
2588 {
2589 "localized_name": "a",
2590 "name": "a",
2591 "shape": 7,
2592 "type": "FLOAT,INT",
2593 "link": 1125
2594 },
2595 {
2596 "localized_name": "b",
2597 "name": "b",
2598 "shape": 7,
2599 "type": "FLOAT,INT",
2600 "link": 1126
2601 },
2602 {
2603 "localized_name": "c",
2604 "name": "c",
2605 "shape": 7,
2606 "type": "FLOAT,INT",
2607 "link": null
2608 },
2609 {
2610 "localized_name": "expression",
2611 "name": "expression",
2612 "type": "STRING",
2613 "widget": {
2614 "name": "expression"
2615 },
2616 "link": null
2617 }
2618 ],
2619 "outputs": [
2620 {
2621 "localized_name": "float",
2622 "name": "float",
2623 "type": "FLOAT",
2624 "links": null
2625 },
2626 {
2627 "localized_name": "int",
2628 "name": "int",
2629 "type": "INT",
2630 "links": [
2631 1129,
2632 1131
2633 ]
2634 }
2635 ],
2636 "title": "Width at 4 or 16 MP",
2637 "properties": {
2638 "Node name for S&R": "AUSBOSS_NODES_MathExpression"
2639 },
2640 "widgets_values": [
2641 "round(a * sqrt(16777216 / (a * b)) / 32) * 32"
2642 ],
2643 "widgets_values_named": {
2644 "expression": "round(a * sqrt(16777216 / (a * b)) / 32) * 32"
2645 },
2646 "color": "#14424d",
2647 "bgcolor": "#161f21"
2648 },
2649 {
2650 "id": 309,
2651 "type": "AUSBOSS_NODES_MathExpression",
2652 "pos": [
2653 1184,
2654 2290
2655 ],
2656 "size": [
2657 300,
2658 134
2659 ],
2660 "flags": {
2661 "collapsed": true
2662 },
2663 "order": 25,
2664 "mode": 0,
2665 "inputs": [
2666 {
2667 "localized_name": "a",
2668 "name": "a",
2669 "shape": 7,
2670 "type": "FLOAT,INT",
2671 "link": 1127
2672 },
2673 {
2674 "localized_name": "b",
2675 "name": "b",
2676 "shape": 7,
2677 "type": "FLOAT,INT",
2678 "link": 1128
2679 },
2680 {
2681 "localized_name": "c",
2682 "name": "c",
2683 "shape": 7,
2684 "type": "FLOAT,INT",
2685 "link": null
2686 },
2687 {
2688 "localized_name": "expression",
2689 "name": "expression",
2690 "type": "STRING",
2691 "widget": {
2692 "name": "expression"
2693 },
2694 "link": null
2695 }
2696 ],
2697 "outputs": [
2698 {
2699 "localized_name": "float",
2700 "name": "float",
2701 "type": "FLOAT",
2702 "links": null
2703 },
2704 {
2705 "localized_name": "int",
2706 "name": "int",
2707 "type": "INT",
2708 "links": [
2709 1130,
2710 1132
2711 ]
2712 }
2713 ],
2714 "title": "Height at 4 or 16 MP",
2715 "properties": {
2716 "Node name for S&R": "AUSBOSS_NODES_MathExpression"
2717 },
2718 "widgets_values": [
2719 "round(b * sqrt(16777216 / (a * b)) / 32) * 32"
2720 ],
2721 "widgets_values_named": {
2722 "expression": "round(b * sqrt(16777216 / (a * b)) / 32) * 32"
2723 },
2724 "color": "#14424d",
2725 "bgcolor": "#161f21"
2726 },
2727 {
2728 "id": 69,
2729 "type": "AUSBOSS_NODES_CalloutTile",
2730 "pos": [
2731 -120,
2732 -110
2733 ],
2734 "size": [
2735 270,
2736 96
2737 ],
2738 "flags": {
2739 "pinned": true
2740 },
2741 "order": 8,
2742 "mode": 0,
2743 "inputs": [
2744 {
2745 "localized_name": "text",
2746 "name": "text",
2747 "type": "STRING",
2748 "widget": {
2749 "name": "text"
2750 },
2751 "link": null
2752 }
2753 ],
2754 "outputs": [],
2755 "title": "Callout",
2756 "properties": {
2757 "Node name for S&R": "AUSBOSS_NODES_CalloutTile",
2758 "cnr_id": "ausboss-nodes",
2759 "ausbossCalloutTile": "youtube"
2760 },
2761 "widgets_values": [
2762 "[AusBoss on YouTube](https://www.youtube.com/@ausbossdev)"
2763 ],
2764 "widgets_values_named": {
2765 "text": "[AusBoss on YouTube](https://www.youtube.com/@ausbossdev)"
2766 },
2767 "color": "#07090a",
2768 "bgcolor": "#07090a"
2769 },
2770 {
2771 "id": 310,
2772 "type": "AUSBOSS_NODES_CalloutTile",
2773 "pos": [
2774 170,
2775 -110
2776 ],
2777 "size": [
2778 270,
2779 96
2780 ],
2781 "flags": {
2782 "pinned": true
2783 },
2784 "order": 43,
2785 "mode": 0,
2786 "inputs": [
2787 {
2788 "localized_name": "text",
2789 "name": "text",
2790 "type": "STRING",
2791 "widget": {
2792 "name": "text"
2793 },
2794 "link": null
2795 }
2796 ],
2797 "outputs": [],
2798 "title": "Callout",
2799 "properties": {
2800 "Node name for S&R": "AUSBOSS_NODES_CalloutTile",
2801 "cnr_id": "ausboss-nodes",
2802 "ausbossCalloutTile": "discord"
2803 },
2804 "widgets_values": [
2805 "[Get help + more workflows](https://discord.gg/RwVxY9Fuaj)"
2806 ],
2807 "widgets_values_named": {
2808 "text": "[Get help + more workflows](https://discord.gg/RwVxY9Fuaj)"
2809 },
2810 "color": "#07090a",
2811 "bgcolor": "#07090a"
2812 },
2813 {
2814 "id": 311,
2815 "type": "AUSBOSS_NODES_WorkflowNoteTile",
2816 "pos": [
2817 -120,
2818 1380
2819 ],
2820 "size": [
2821 560,
2822 88
2823 ],
2824 "flags": {
2825 "pinned": true
2826 },
2827 "order": 44,
2828 "mode": 0,
2829 "inputs": [
2830 {
2831 "localized_name": "note",
2832 "name": "note",
2833 "type": "STRING",
2834 "widget": {
2835 "name": "note"
2836 },
2837 "link": null
2838 }
2839 ],
2840 "outputs": [],
2841 "title": "🆎 AusBoss",
2842 "properties": {
2843 "Node name for S&R": "AUSBOSS_NODES_WorkflowNoteTile",
2844 "cnr_id": "ausboss-nodes"
2845 },
2846 "widgets_values": [
2847 "{\"v\": 1, \"title\": \"More pictures\", \"subtitle\": \"Up to seven more pictures to take things from.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- To take something from another picture, plug any image loader into **picture 2** on the box **Write the prompt**, and say \\\"put the hat from <Picture 2> on her\\\". Up to **picture 8**, in order.\\n- Your loaded picture is always Picture 1, the one that gets changed.\", \"models\": [], \"packs\": [], \"links\": []}"
2848 ],
2849 "widgets_values_named": {
2850 "note": "{\"v\": 1, \"title\": \"More pictures\", \"subtitle\": \"Up to seven more pictures to take things from.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- To take something from another picture, plug any image loader into **picture 2** on the box **Write the prompt**, and say \\\"put the hat from <Picture 2> on her\\\". Up to **picture 8**, in order.\\n- Your loaded picture is always Picture 1, the one that gets changed.\", \"models\": [], \"packs\": [], \"links\": []}"
2851 },
2852 "color": "#14424d",
2853 "bgcolor": "#161f21"
2854 },
2855 {
2856 "id": 312,
2857 "type": "AUSBOSS_NODES_WorkflowNoteTile",
2858 "pos": [
2859 -120,
2860 1478
2861 ],
2862 "size": [
2863 560,
2864 88
2865 ],
2866 "flags": {
2867 "pinned": true
2868 },
2869 "order": 44,
2870 "mode": 0,
2871 "inputs": [
2872 {
2873 "localized_name": "note",
2874 "name": "note",
2875 "type": "STRING",
2876 "widget": {
2877 "name": "note"
2878 },
2879 "link": null
2880 }
2881 ],
2882 "outputs": [],
2883 "title": "🆎 AusBoss",
2884 "properties": {
2885 "Node name for S&R": "AUSBOSS_NODES_WorkflowNoteTile",
2886 "cnr_id": "ausboss-nodes"
2887 },
2888 "widgets_values": [
2889 "{\"v\": 1, \"title\": \"Tips\", \"subtitle\": \"Prompts, edits and LoRAs.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- Switch **Auto prompt** on for short prompts. Without it they come out plain, and any words in them come out garbled. With it, posters, covers and infographics spell their words right.\\n- On an edit, **Auto prompt** makes a bigger change from a few words (\\\"vampire\\\", \\\"winter\\\"). For a small, exact change, leave it off and say what should stay the same.\\n- Leave **Auto prompt** off when your prompt is already a full description. In our runs of the opening prompt, rewriting it tangled the hands on the can.\\n- Put any words that should appear in quotes, and say where they go.\\n- For an edit, say what changes and what stays: \\\"Change her top to a black varsity jacket. Keep everything else the same.\\\"\\n- The model makes 6 steps with no CFG. Don't add steps, CFG or a negative prompt: it is made for this schedule.\\n- Your LoRAs go in the **LoRA Loader**, below the locked Fast row.\\n- If the run stops with \\\"Value not in list\\\", a file is missing. Download it from the list below.\", \"models\": [], \"packs\": [], \"links\": []}"
2890 ],
2891 "widgets_values_named": {
2892 "note": "{\"v\": 1, \"title\": \"Tips\", \"subtitle\": \"Prompts, edits and LoRAs.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- Switch **Auto prompt** on for short prompts. Without it they come out plain, and any words in them come out garbled. With it, posters, covers and infographics spell their words right.\\n- On an edit, **Auto prompt** makes a bigger change from a few words (\\\"vampire\\\", \\\"winter\\\"). For a small, exact change, leave it off and say what should stay the same.\\n- Leave **Auto prompt** off when your prompt is already a full description. In our runs of the opening prompt, rewriting it tangled the hands on the can.\\n- Put any words that should appear in quotes, and say where they go.\\n- For an edit, say what changes and what stays: \\\"Change her top to a black varsity jacket. Keep everything else the same.\\\"\\n- The model makes 6 steps with no CFG. Don't add steps, CFG or a negative prompt: it is made for this schedule.\\n- Your LoRAs go in the **LoRA Loader**, below the locked Fast row.\\n- If the run stops with \\\"Value not in list\\\", a file is missing. Download it from the list below.\", \"models\": [], \"packs\": [], \"links\": []}"
2893 },
2894 "color": "#14424d",
2895 "bgcolor": "#161f21"
2896 },
2897 {
2898 "id": 313,
2899 "type": "AUSBOSS_NODES_WorkflowNoteTile",
2900 "pos": [
2901 -120,
2902 1576
2903 ],
2904 "size": [
2905 560,
2906 88
2907 ],
2908 "flags": {
2909 "pinned": true
2910 },
2911 "order": 44,
2912 "mode": 0,
2913 "inputs": [
2914 {
2915 "localized_name": "note",
2916 "name": "note",
2917 "type": "STRING",
2918 "widget": {
2919 "name": "note"
2920 },
2921 "link": null
2922 }
2923 ],
2924 "outputs": [],
2925 "title": "🆎 AusBoss",
2926 "properties": {
2927 "Node name for S&R": "AUSBOSS_NODES_WorkflowNoteTile",
2928 "cnr_id": "ausboss-nodes"
2929 },
2930 "widgets_values": [
2931 "{\"v\": 1, \"title\": \"What it is bad at\", \"subtitle\": \"Small far-off text, very short prompts.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- Small text far from the subject (fine print, credits, labels in the background) can still come out as nonsense.\\n- Short prompts with Auto prompt off can invent filler words.\", \"models\": [], \"packs\": [], \"links\": []}"
2932 ],
2933 "widgets_values_named": {
2934 "note": "{\"v\": 1, \"title\": \"What it is bad at\", \"subtitle\": \"Small far-off text, very short prompts.\", \"author\": \"AusBoss\", \"layout\": \"tile\", \"accent\": \"\", \"body\": \"- Small text far from the subject (fine print, credits, labels in the background) can still come out as nonsense.\\n- Short prompts with Auto prompt off can invent filler words.\", \"models\": [], \"packs\": [], \"links\": []}"
2935 },
2936 "color": "#14424d",
2937 "bgcolor": "#161f21"
2938 }
2939 ],
2940 "links": [
2941 [
2942 1000,
2943 3,
2944 0,
2945 41,
2946 0,
2947 "STRING"
2948 ],
2949 [
2950 1001,
2951 6,
2952 0,
2953 41,
2954 8,
2955 "IMAGE"
2956 ],
2957 [
2958 1002,
2959 70,
2960 0,
2961 41,
2962 9,
2963 "CLIP"
2964 ],
2965 [
2966 1005,
2967 41,
2968 0,
2969 15,
2970 0,
2971 "STRING"
2972 ],
2973 [
2974 1006,
2975 41,
2976 0,
2977 39,
2978 0,
2979 "STRING"
2980 ],
2981 [
2982 1007,
2983 41,
2984 1,
2985 39,
2986 8,
2987 "IMAGE"
2988 ],
2989 [
2990 1008,
2991 41,
2992 2,
2993 39,
2994 9,
2995 "IMAGE"
2996 ],
2997 [
2998 1009,
2999 41,
3000 3,
3001 39,
3002 10,
3003 "IMAGE"
3004 ],
3005 [
3006 1010,
3007 41,
3008 4,
3009 39,
3010 11,
3011 "IMAGE"
3012 ],
3013 [
3014 1011,
3015 41,
3016 5,
3017 39,
3018 12,
3019 "IMAGE"
3020 ],
3021 [
3022 1012,
3023 41,
3024 6,
3025 39,
3026 13,
3027 "IMAGE"
3028 ],
3029 [
3030 1013,
3031 41,
3032 7,
3033 39,
3034 14,
3035 "IMAGE"
3036 ],
3037 [
3038 1014,
3039 4,
3040 0,
3041 39,
3042 1,
3043 "INT"
3044 ],
3045 [
3046 1015,
3047 6,
3048 0,
3049 39,
3050 2,
3051 "IMAGE"
3052 ],
3053 [
3054 1018,
3055 8,
3056 0,
3057 39,
3058 5,
3059 "CLIP"
3060 ],
3061 [
3062 1019,
3063 9,
3064 0,
3065 39,
3066 6,
3067 "VAE"
3068 ],
3069 [
3070 1020,
3071 10,
3072 0,
3073 11,
3074 0,
3075 "MODEL"
3076 ],
3077 [
3078 1021,
3079 11,
3080 0,
3081 39,
3082 7,
3083 "MODEL"
3084 ],
3085 [
3086 1022,
3087 39,
3088 0,
3089 105,
3090 0,
3091 "IMAGE"
3092 ],
3093 [
3094 1023,
3095 4,
3096 0,
3097 105,
3098 1,
3099 "INT"
3100 ],
3101 [
3102 1024,
3103 106,
3104 0,
3105 105,
3106 2,
3107 "MODEL"
3108 ],
3109 [
3110 1025,
3111 107,
3112 0,
3113 105,
3114 3,
3115 "VAE"
3116 ],
3117 [
3118 1026,
3119 105,
3120 0,
3121 13,
3122 0,
3123 "IMAGE"
3124 ],
3125 [
3126 1027,
3127 39,
3128 0,
3129 12,
3130 0,
3131 "IMAGE"
3132 ],
3133 [
3134 1028,
3135 105,
3136 0,
3137 12,
3138 1,
3139 "IMAGE"
3140 ],
3141 [
3142 1029,
3143 39,
3144 1,
3145 54,
3146 0,
3147 "IMAGE"
3148 ],
3149 [
3150 1030,
3151 39,
3152 0,
3153 54,
3154 1,
3155 "IMAGE"
3156 ],
3157 [
3158 1125,
3159 5,
3160 0,
3161 308,
3162 0,
3163 "INT"
3164 ],
3165 [
3166 1126,
3167 5,
3168 1,
3169 308,
3170 1,
3171 "INT"
3172 ],
3173 [
3174 1127,
3175 5,
3176 0,
3177 309,
3178 0,
3179 "INT"
3180 ],
3181 [
3182 1128,
3183 5,
3184 1,
3185 309,
3186 1,
3187 "INT"
3188 ],
3189 [
3190 1129,
3191 308,
3192 1,
3193 41,
3194 10,
3195 "INT"
3196 ],
3197 [
3198 1130,
3199 309,
3200 1,
3201 41,
3202 11,
3203 "INT"
3204 ],
3205 [
3206 1131,
3207 308,
3208 1,
3209 39,
3210 3,
3211 "INT"
3212 ],
3213 [
3214 1132,
3215 309,
3216 1,
3217 39,
3218 4,
3219 "INT"
3220 ]
3221 ],
3222 "groups": [
3223 {
3224 "id": 1,
3225 "title": "1 · Your input",
3226 "bounding": [
3227 480,
3228 0,
3229 928,
3230 2480
3231 ],
3232 "color": "#1d5560",
3233 "flags": {
3234 "pinned": true
3235 }
3236 },
3237 {
3238 "id": 2,
3239 "title": "2 · Behind the scenes",
3240 "bounding": [
3241 1448,
3242 0,
3243 1260,
3244 1090
3245 ],
3246 "color": "#1d5560",
3247 "flags": {
3248 "pinned": true
3249 }
3250 },
3251 {
3252 "id": 3,
3253 "title": "3 · Models",
3254 "bounding": [
3255 1448,
3256 1090,
3257 960,
3258 700
3259 ],
3260 "color": "#1d5560",
3261 "flags": {
3262 "pinned": true
3263 }
3264 },
3265 {
3266 "id": 4,
3267 "title": "4 · Result",
3268 "bounding": [
3269 2748,
3270 0,
3271 500,
3272 2180
3273 ],
3274 "color": "#1d5560",
3275 "flags": {
3276 "pinned": true
3277 }
3278 }
3279 ],
3280 "definitions": {
3281 "subgraphs": [
3282 {
3283 "id": "763be216-d720-52d4-b604-469834ebd74c",
3284 "version": 1,
3285 "state": {
3286 "lastGroupId": 22,
3287 "lastNodeId": 470,
3288 "lastLinkId": 1132,
3289 "lastRerouteId": 0
3290 },
3291 "revision": 0,
3292 "config": [],
3293 "name": "Write the prompt",
3294 "inputNode": {
3295 "id": -10,
3296 "bounding": [
3297 -260,
3298 120,
3299 150,
3300 320
3301 ]
3302 },
3303 "outputNode": {
3304 "id": -20,
3305 "bounding": [
3306 2210,
3307 120,
3308 150,
3309 200
3310 ]
3311 },
3312 "inputs": [
3313 {
3314 "id": "8d8427cb-90dd-5fa9-b01c-122f97e7d0e8",
3315 "name": "your prompt",
3316 "type": "*",
3317 "linkIds": [
3318 1033,
3319 1047,
3320 1144
3321 ],
3322 "localized_name": "your prompt",
3323 "pos": [
3324 51,
3325 24
3326 ]
3327 },
3328 {
3329 "id": "b7201a4f-4e43-5199-a8d1-7e380838d520",
3330 "name": "picture 2",
3331 "type": "IMAGE",
3332 "linkIds": [
3333 1036,
3334 1057
3335 ],
3336 "localized_name": "picture 2",
3337 "shape": 7,
3338 "pos": [
3339 51,
3340 44
3341 ]
3342 },
3343 {
3344 "id": "5b9da575-ec4f-5d79-8b38-7fcdde0abb0e",
3345 "name": "picture 3",
3346 "type": "IMAGE",
3347 "linkIds": [
3348 1037,
3349 1059
3350 ],
3351 "localized_name": "picture 3",
3352 "shape": 7,
3353 "pos": [
3354 51,
3355 64
3356 ]
3357 },
3358 {
3359 "id": "ac084225-d7d4-530a-9f0f-bc0942faf396",
3360 "name": "picture 4",
3361 "type": "IMAGE",
3362 "linkIds": [
3363 1038,
3364 1061
3365 ],
3366 "localized_name": "picture 4",
3367 "shape": 7,
3368 "pos": [
3369 51,
3370 84
3371 ]
3372 },
3373 {
3374 "id": "39f32069-d0ac-5732-8f8c-ed7ac61aa29a",
3375 "name": "picture 5",
3376 "type": "IMAGE",
3377 "linkIds": [
3378 1039,
3379 1063
3380 ],
3381 "localized_name": "picture 5",
3382 "shape": 7,
3383 "pos": [
3384 51,
3385 104
3386 ]
3387 },
3388 {
3389 "id": "f9b23622-da24-51e4-a999-61dee5f6a271",
3390 "name": "picture 6",
3391 "type": "IMAGE",
3392 "linkIds": [
3393 1040,
3394 1065
3395 ],
3396 "localized_name": "picture 6",
3397 "shape": 7,
3398 "pos": [
3399 51,
3400 124
3401 ]
3402 },
3403 {
3404 "id": "ec278436-aad6-53ce-894b-dbed078229f9",
3405 "name": "picture 7",
3406 "type": "IMAGE",
3407 "linkIds": [
3408 1041,
3409 1067
3410 ],
3411 "localized_name": "picture 7",
3412 "shape": 7,
3413 "pos": [
3414 51,
3415 144
3416 ]
3417 },
3418 {
3419 "id": "c479187a-b189-5123-bd39-eb3336f3839b",
3420 "name": "picture 8",
3421 "type": "IMAGE",
3422 "linkIds": [
3423 1042,
3424 1069
3425 ],
3426 "localized_name": "picture 8",
3427 "shape": 7,
3428 "pos": [
3429 51,
3430 164
3431 ]
3432 },
3433 {
3434 "id": "f3b82cc3-3d99-58bc-9493-f36c3137f5ca",
3435 "name": "your picture",
3436 "type": "IMAGE",
3437 "linkIds": [
3438 1034
3439 ],
3440 "localized_name": "your picture",
3441 "shape": 7,
3442 "pos": [
3443 51,
3444 184
3445 ]
3446 },
3447 {
3448 "id": "3d179a3d-0152-5b63-b444-34512f766257",
3449 "name": "prompt writer",
3450 "type": "CLIP",
3451 "linkIds": [
3452 1142
3453 ],
3454 "localized_name": "prompt writer",
3455 "pos": [
3456 51,
3457 204
3458 ]
3459 },
3460 {
3461 "id": "a604fbce-2616-5a83-a19c-a03ef1d57bee",
3462 "name": "width",
3463 "type": "INT",
3464 "linkIds": [
3465 1031
3466 ],
3467 "localized_name": "width",
3468 "shape": 7,
3469 "pos": [
3470 51,
3471 224
3472 ]
3473 },
3474 {
3475 "id": "83563c3f-c0f2-510f-855a-5a078d594519",
3476 "name": "height",
3477 "type": "INT",
3478 "linkIds": [
3479 1032
3480 ],
3481 "localized_name": "height",
3482 "shape": 7,
3483 "pos": [
3484 51,
3485 244
3486 ]
3487 },
3488 {
3489 "id": "f9e0c53c-1284-56dc-a625-d45a0b1854f9",
3490 "name": "write an edit",
3491 "type": "BOOLEAN",
3492 "linkIds": [
3493 1055,
3494 1071
3495 ],
3496 "localized_name": "write an edit",
3497 "pos": [
3498 51,
3499 264
3500 ]
3501 },
3502 {
3503 "id": "a383939d-c11f-4395-a7c5-12c299472db2",
3504 "name": "auto prompt",
3505 "type": "BOOLEAN",
3506 "linkIds": [
3507 1145
3508 ],
3509 "localized_name": "auto prompt",
3510 "pos": [
3511 51,
3512 284
3513 ]
3514 }
3515 ],
3516 "outputs": [
3517 {
3518 "id": "a396e8d5-70dd-5a84-ab50-949682137cd7",
3519 "name": "prompt",
3520 "type": "STRING",
3521 "linkIds": [
3522 1056
3523 ],
3524 "localized_name": "prompt",
3525 "pos": [
3526 24,
3527 24
3528 ]
3529 },
3530 {
3531 "id": "01ca7308-80f9-5cbc-9c34-069e98802478",
3532 "name": "picture 2",
3533 "type": "IMAGE",
3534 "linkIds": [
3535 1058
3536 ],
3537 "localized_name": "picture 2",
3538 "pos": [
3539 24,
3540 44
3541 ]
3542 },
3543 {
3544 "id": "6634f4ee-c290-53b7-bc46-90c29d2c4883",
3545 "name": "picture 3",
3546 "type": "IMAGE",
3547 "linkIds": [
3548 1060
3549 ],
3550 "localized_name": "picture 3",
3551 "pos": [
3552 24,
3553 64
3554 ]
3555 },
3556 {
3557 "id": "67901101-da79-5a21-b20b-06c2572f77b9",
3558 "name": "picture 4",
3559 "type": "IMAGE",
3560 "linkIds": [
3561 1062
3562 ],
3563 "localized_name": "picture 4",
3564 "pos": [
3565 24,
3566 84
3567 ]
3568 },
3569 {
3570 "id": "c589663f-cb2b-5a0f-b076-0aa38caf6920",
3571 "name": "picture 5",
3572 "type": "IMAGE",
3573 "linkIds": [
3574 1064
3575 ],
3576 "localized_name": "picture 5",
3577 "pos": [
3578 24,
3579 104
3580 ]
3581 },
3582 {
3583 "id": "697bf7a5-c939-54e4-ac42-0437f3d4ecde",
3584 "name": "picture 6",
3585 "type": "IMAGE",
3586 "linkIds": [
3587 1066
3588 ],
3589 "localized_name": "picture 6",
3590 "pos": [
3591 24,
3592 124
3593 ]
3594 },
3595 {
3596 "id": "8c02be2a-3f21-5b37-ad04-87dc1ac0a3ee",
3597 "name": "picture 7",
3598 "type": "IMAGE",
3599 "linkIds": [
3600 1068
3601 ],
3602 "localized_name": "picture 7",
3603 "pos": [
3604 24,
3605 144
3606 ]
3607 },
3608 {
3609 "id": "bf5384b8-f671-5af9-9420-2a6997333ec5",
3610 "name": "picture 8",
3611 "type": "IMAGE",
3612 "linkIds": [
3613 1070
3614 ],
3615 "localized_name": "picture 8",
3616 "pos": [
3617 24,
3618 164
3619 ]
3620 }
3621 ],
3622 "widgets": [],
3623 "nodes": [
3624 {
3625 "id": 401,
3626 "type": "AUSBOSS_NODES_Callout",
3627 "pos": [
3628 20,
3629 80
3630 ],
3631 "size": [
3632 460,
3633 150
3634 ],
3635 "flags": {
3636 "pinned": true
3637 },
3638 "order": 0,
3639 "mode": 0,
3640 "inputs": [
3641 {
3642 "localized_name": "text",
3643 "name": "text",
3644 "type": "STRING",
3645 "widget": {
3646 "name": "text"
3647 },
3648 "link": null
3649 }
3650 ],
3651 "outputs": [],
3652 "title": "Callout",
3653 "properties": {
3654 "Node name for S&R": "AUSBOSS_NODES_Callout",
3655 "cnr_id": "ausboss-nodes"
3656 },
3657 "widgets_values": [
3658 "For a new picture, the writer gets MiniMax's own prompt-expander instructions, your words and the picture's shape.\n###### It writes one long English description of the whole picture: the prompt MiniMax-H3-Image is made for."
3659 ],
3660 "widgets_values_named": {
3661 "text": "For a new picture, the writer gets MiniMax's own prompt-expander instructions, your words and the picture's shape.\n###### It writes one long English description of the whole picture: the prompt MiniMax-H3-Image is made for."
3662 },
3663 "color": "#07090a",
3664 "bgcolor": "#07090a"
3665 },
3666 {
3667 "id": 402,
3668 "type": "StringFormat",
3669 "pos": [
3670 20,
3671 474
3672 ],
3673 "size": [
3674 460,
3675 170
3676 ],
3677 "flags": [],
3678 "order": 5,
3679 "mode": 0,
3680 "inputs": [
3681 {
3682 "localized_name": "a",
3683 "name": "values.a",
3684 "shape": 7,
3685 "type": "*",
3686 "link": 1073
3687 },
3688 {
3689 "localized_name": "b",
3690 "name": "values.b",
3691 "shape": 7,
3692 "type": "*",
3693 "link": null
3694 },
3695 {
3696 "localized_name": "f_string",
3697 "name": "f_string",
3698 "type": "STRING",
3699 "widget": {
3700 "name": "f_string"
3701 },
3702 "link": null
3703 },
3704 {
3705 "localized_name": "c",
3706 "name": "values.c",
3707 "shape": 7,
3708 "type": "*",
3709 "link": 1033
3710 },
3711 {
3712 "localized_name": "d",
3713 "name": "values.d",
3714 "shape": 7,
3715 "type": "*",
3716 "link": null
3717 }
3718 ],
3719 "outputs": [
3720 {
3721 "localized_name": "STRING",
3722 "name": "STRING",
3723 "type": "STRING",
3724 "links": [
3725 1050
3726 ]
3727 }
3728 ],
3729 "title": "Format Text - MiniMax's expander instructions, your words, the shape",
3730 "properties": {
3731 "Node name for S&R": "StringFormat",
3732 "cnr_id": "comfy-core",
3733 "ver": "0.44.0"
3734 },
3735 "widgets_values": [
3736 "You are a caption-expansion and canvas-selection model for still-image\ntext-to-image generation. Rewrite input_text into a natural English detailed\ncaption and choose one aspect ratio for the requested final image. After private\nreasoning, serialize the complete caption and aspect ratio together as the one\nrequired two-key JSON object; no visible analysis, markdown, or self-talk. The\ncaption may be long when the requested image is a complex information or design artifact.\n\nRole: act like an expert visual designer and technical illustrator, not a word\nexpander. Preserve every hard fact from the input, then complete the image with\nvisible, inspectable content. The goal is not more words; the goal is that a\nviewer can inspect each important part of the image and find meaningful subjects,\nlabels, relationships, materials, data, and hierarchy.\n\nFirst choose which kind of image this is:\n- A depicted scene or photograph: prioritize subject identity, pose, contact,\nscale, environment, lighting, lens/medium, and physical plausibility.\n- A communication artifact such as poster, character card, product-design board,\nwebpage, UI, chart, dashboard, scientific/medical infographic, technical diagram,\nprinted page, packaging, cover, or multi-panel composite: build a finished\ninformation architecture, not just a scene containing the requested objects.\n\nAspect-ratio decision policy:\nChoose exactly one of these allowed ratios: 1:1, 5:4, 4:3, 3:2, 16:9, 21:9,\n4:5, 3:4, 2:3, or 9:16. Decide in the following order; earlier evidence\noverrides later evidence.\n\n1. Explicit final canvas\n- Preserve an explicit final-output ratio. Map explicit output pixel dimensions\n by width divided by height to the closest allowed ratio.\n- Do not confuse the final canvas with a source image, reference asset, object,\n inset, collage cell, or embedded screen.\n\n2. Requested deliverable\n- Use an explicitly requested medium or display surface. Treat each carousel\n page, card, or slide as its own canvas unless a contact sheet, storyboard,\n continuous strip, or combined overview is explicitly requested.\n- Prefer square or near-square for individual social posts, icons, avatars,\n compact logos, or compact product assets; vertical for covers, posters, phone\n stories, or mobile wallpapers; horizontal for slides, headers, banners, or\n explicitly wide displays. Do not invent a platform or deliverable.\n\n3. Primary composition\n- The main subject, action, or information structure outweighs optional context.\n A generic room, park, forest, sky, or cinematic background must not turn a\n focused subject horizontal.\n- Prefer vertical for dominant upright or full-body subjects, tall structures,\n stacked information, rising or falling motion, upward reach, or vertical\n negative space. Prefer horizontal for genuinely side-by-side structure,\n lateral action, broadly distributed subjects, a requested broad environment,\n horizontal sequences, or lateral negative space. Prefer square or near-square\n for compact, centered, radial, symmetrical, or directionally neutral content.\n\n4. Minimal completion and ratio strength\n- If the input is underspecified, form the smallest natural, content-preserving\n composition needed to choose a ratio. Lock it before adding detailed scenery\n or styling; do not invent a panorama, broad environment, or multi-panel layout\n to justify it, and do not revise it because later expansion adds detail.\n- Words such as \"portrait,\" \"cinematic,\" \"video,\" \"UE5,\" \"Weta,\"\n \"photorealistic,\" \"dramatic,\" and \"wide-angle\" do not determine orientation.\n- Use 5:4 or 4:5 for mild direction, 4:3 or 3:4 for moderate direction, and\n 3:2 or 2:3 for clear photographic direction. Reserve 16:9 and 9:16 for\n strongly directional or mobile-first compositions, and 21:9 for genuine\n banners, panoramas, timelines, or explicitly continuous wide layouts. When\n evidence remains weak, prefer 1:1 or a moderate ratio.\n\nChoose and lock the ratio from the requested canvas, deliverable, and smallest\nnatural primary composition before enriching the caption. Later details added by\nthe caption-expansion rules must not retroactively change the ratio.\n\nUniversal hard rules:\n1. Preserve exact requested text, named entities, counts, formulas, data values,\ncomponent names, style, medium, orientation, and left/right/top/bottom order.\n2. Required content must be visible, not implied. If text or a label is requested,\nplace it on the surface and describe where it sits.\n3. For active bodies, solve scene geometry before pose detail. Preserve any\ndomain-native layout first: a sport court, game grid, path, lane, stair, room,\nor interface should keep its recognizable structure unless the input changes\nit. Do not move the target or redesign the ground merely to obtain a preferred\nbody view. Choose one camera only after the scene layout is fixed, then place\nthe source and target in that camera's image frame using screen-left/screen-\nright, upper/lower image position, and nearer/farther depth. Silently trace the\nsource-to-target vector in this fixed view. Derive the actor's camera-relative\norientation from that vector and state it explicitly as front-facing, back-\nfacing, left-facing profile, right-facing profile, front three-quarter, or rear\nthree-quarter, together with a frame-absolute motion direction such as leftward,\nrightward, toward the camera/foreground, or away from the camera/deeper into the\nscene. Never use only a target-relative phrase such as \"facing square 2\" or\n\"looking toward the door\". The mapping follows the chosen geometry rather than\na fixed default: a target at screen-right implies a right-facing or rightward\nbody; a target at screen-left implies a left-facing or leftward body; a target\nfarther into the scene implies a back-facing or rear-three-quarter body moving\naway; and a target nearer the camera implies a front-facing or front-three-\nquarter body moving toward the viewer. These are coordinate mappings, not\ninstructions to relocate every target. Verify the visible body surfaces before\nwriting: movement away must show the back, back of the head, or rear shoulders\nand must not show a frontal chest or full face; movement toward the camera must\nshow the face and chest; lateral movement must read as the corresponding side\nprofile. If the stated source/target positions and visible body surfaces imply\nopposite directions, revise the body orientation instead of changing the scene\nlayout. Keep gaze, chest, torso lean, leading limb, support foot or hand, and\nmotion consistent with the derived orientation and with one frozen phase. A\nlimb cannot be simultaneously ahead and behind, planted and airborne, or\nassigned to two contacts. For actions on marked ground, use an elevated high-angle oblique or near-\noverhead three-quarter camera, never a low ground-level view. Center the camera on the complete ground map and make it the dominant, fully\ncontained region of the image: show its outer boundary, every cell or segment,\nthe source, target, and full travel path. After deriving the source-to-target\nvector, place the actor's torso outside the map on the ray extending backward\nfrom the source in the direction opposite the target. Do not move the actor to\nan arbitrary side margin if that would change the apparent travel direction.\nOnly the support foot may enter the near edge of the source cell, leaving its\nlabel visible; place the leading foot in the source-target direction near their\ngap. Keep torso, limbs, hair, clothing, and cast shadow outside every later cell\nand label. Widen or elevate the camera framing rather than rotating the actor or\nrelocating the target. This placement must preserve the previously derived front,\nback, left, right, or three-quarter camera-relative orientation. Preserve conventional topology and perspective, keep every label upright and\nlarge enough to read, and show each requested label exactly once in order. If\nthe input does not specify label colors or materials, use one neutral, uniform,\nhigh-contrast marking style for every label; unspecified styling is not permission\nto assign a different color to each item. Keep the actor's hands empty unless the\ninput names a held object. Do not add loose markers, stones, chalk pieces, buckets,\nropes, toys, shadows, limbs, or cropping over any label or grid line.\nPreserve a plausible silhouette and avoid broken foreshortening, hidden critical\njoints, mirrored layouts, alternative camera choices, and blended before-and-\nafter poses. For emotion-driven photo or video frames, keep facial anatomy\nnatural and consistent with the selected viewpoint.\n4. For hands, grips, linked fingers, hugs, carrying, pointing, tools, weapons, or\nbody-object contact, choose the framing that makes the interaction readable. Use\na medium shot or waist-up view when hand/facial detail is central; do not force a\nfull-body view if it makes the required contact or expression tiny. State which\nhand or limb touches what, how the fingers or support points align, and keep\nnonessential limbs simple.\n5. Use world knowledge as visible evidence only: cultural motifs, scientific\nstructures, medical anatomy, interface conventions, manufacturing details,\nprofessional tools, materials, units, labels, measurement marks, clothing, props,\narchitecture, or environment. Do not add lore that cannot be seen.\n6. Respect requested subject count. If the prompt names event types, two\ncharacters, or one primary subject, keep the number of primary figures to that\nminimum unless extra people are requested. Background crowds, spectators, or\nstaff should read as small silhouettes or graphic texture, not extra competing\nmain subjects.\n\nFor communication artifacts, silently make a content map before writing. Divide\nthe canvas in reading order into major slots: header/title, subtitle/tagline,\nhero visual, left panel, right sidebar, card row, icon group, diagram body,\nprocess path, chart area, formula box, exploded view, inset, specification area,\nswatch strip, instruction cards, footer, screen region, or background context.\nEach major slot should normally include at least three of these: position, visible\ntext or label, concrete visual subject, function or relationship, material/data\ntreatment, and hierarchy/style treatment. Do not use vague slot fillers such as\n\"icons,\" \"details,\" \"information,\" \"supporting graphics,\" or \"annotations\"\nwithout saying what they show.\n\nMake the artifact feel complete. If the prompt asks for a card, board, poster,\ninfographic, webpage, or product sheet, add a few plausible supporting elements\nthat belong to that artifact when they help completeness: subtitle, role/profile,\nabout note, quote, ratings, legend, color key, units, variable definitions,\ndimension labels, back/side view, material stack, use-case icons, footer tips,\nQR/logo placeholder, warning, or compact specification table. Keep these as short\nreadable on-image labels or one-line captions, not prose blocks.\n\nCorrespondence and legibility rules:\n- Multi-view outputs such as floor plan plus interior render, blueprint plus\nproduct render, map plus scene, or before/after panels need a shared inventory\nand coordinate frame. Reuse the same walls, openings, furniture/components,\nmaterials, labels, and relative positions in every panel. Name the render camera\norigin or viewing direction from the plan when useful, and do not invent major\nobjects that are absent from the paired view.\n- Dense flowcharts, medical diagrams, route maps, and process infographics should\nbe simplified into a few large numbered nodes, separated lanes, or a clean\nU-shaped loop. Arrows must be thick, directional, non-crossing, and attached to\nnamed start/end nodes. Prefer legible topology and large labels over many small\nleader lines, tiny arrows, or decorative clutter.\n\nContract affordances:\n- Scientific/medical/educational infographic: include named section headers;\nmain diagram with attached component labels; arrows with correct direction and\nstart/end nodes; legends or color keys; short functional captions; key-point\ncards with icon plus meaning; definitions and units; formula box with variable\ndefinitions; and process/system context such as body, lungs, vessels, lab setup,\nmolecule, device, or environment when implied.\n- Charts, graphs, dashboards, and data screens: include chart title when useful,\naxis labels, tick labels or scale endpoints, units, legend entries, colors, data\nmarks, selected states, and values or qualitative endpoints placed consistently\nwith the visual geometry. Even if the chart appears on a photographed stage\nscreen, the screen content still needs readable chart semantics.\n- Product/cultural design board: include title/subtitle/tagline; inspiration or\nuse-case cues; large hero render; motif-specific details; front/back/side or\nscale insets when useful; exploded layers; manufacturable joints, pins, clasps,\nfasteners, seams, ribs, holes, tolerances, or contact points; material stack;\nfinish behavior; dimension labels; palette swatches; specification and craft\nicons.\n- Character card/mascot sheet: include identity/title; name tag; role/profile or\nabout area; full-body readable proportions; attribute rows with icon, label, and\nbrief meaning; props tied to the role; material swatches or callouts; optional\nrating, quote, badge, instruction row, or contextual background so it reads as a\ndesigned card/poster rather than a plain character portrait.\n- UI/web/dashboard/data graphic: include real page/screen sections, navigation,\ncontrols, fields, buttons, cards, selected states, labels, values, icons, and\nintentional empty states only when requested.\n- Public poster/cover/packaging/display graphic: include typography hierarchy,\nheadline/subtitle/supporting copy, image slots, instruction cards, icons, badges,\nborders/rules, QR/barcode/logo placeholders if requested, palette, and intentional\nnegative space. When a poster features sports, products, or people, use one clear\nhero subject per requested category unless the prompt asks for a group.\n\nStyle comes after content. Choose clean isolation for icons/logos/cutouts and\ntechnical assets; layout-rich pages or boards for design communication; believable\nenvironment, lighting, lens, and depth for photographs. Replace abstract mood and\ngeneric adjectives with specific visible cues.\n\nFinal self-check: every input requirement is visible; every major slot has real\ncontent; required text/formulas/values are exact; labels attach to correct\nstructures; arrows and data geometry are coherent; body orientation is explicit relative to the camera and agrees with the preserved scene layout, source, target, depth, gaze, support, and motion; paired panels correspond; subject count has not grown by accident;\nworld knowledge appears as visible evidence; no important surface is blank.\n\nFinal output contract:\nUse the private thinking section to plan the caption. The final canvas aspect ratio is already fixed by the user and is given after the input; compose for it and do not choose another. After thinking, write only the caption: one fluent, detailed English paragraph that follows every rule above. No JSON, no labels, no Markdown, no lists, no alternatives, and no text before or after the caption.\n\ninput_text: {c}\nFinal canvas shape: width divided by height is {a:.3g} (below 1 is portrait, above 1 is landscape)\noutput_text:\n"
3737 ],
3738 "widgets_values_named": {
3739 "f_string": "You are a caption-expansion and canvas-selection model for still-image\ntext-to-image generation. Rewrite input_text into a natural English detailed\ncaption and choose one aspect ratio for the requested final image. After private\nreasoning, serialize the complete caption and aspect ratio together as the one\nrequired two-key JSON object; no visible analysis, markdown, or self-talk. The\ncaption may be long when the requested image is a complex information or design artifact.\n\nRole: act like an expert visual designer and technical illustrator, not a word\nexpander. Preserve every hard fact from the input, then complete the image with\nvisible, inspectable content. The goal is not more words; the goal is that a\nviewer can inspect each important part of the image and find meaningful subjects,\nlabels, relationships, materials, data, and hierarchy.\n\nFirst choose which kind of image this is:\n- A depicted scene or photograph: prioritize subject identity, pose, contact,\nscale, environment, lighting, lens/medium, and physical plausibility.\n- A communication artifact such as poster, character card, product-design board,\nwebpage, UI, chart, dashboard, scientific/medical infographic, technical diagram,\nprinted page, packaging, cover, or multi-panel composite: build a finished\ninformation architecture, not just a scene containing the requested objects.\n\nAspect-ratio decision policy:\nChoose exactly one of these allowed ratios: 1:1, 5:4, 4:3, 3:2, 16:9, 21:9,\n4:5, 3:4, 2:3, or 9:16. Decide in the following order; earlier evidence\noverrides later evidence.\n\n1. Explicit final canvas\n- Preserve an explicit final-output ratio. Map explicit output pixel dimensions\n by width divided by height to the closest allowed ratio.\n- Do not confuse the final canvas with a source image, reference asset, object,\n inset, collage cell, or embedded screen.\n\n2. Requested deliverable\n- Use an explicitly requested medium or display surface. Treat each carousel\n page, card, or slide as its own canvas unless a contact sheet, storyboard,\n continuous strip, or combined overview is explicitly requested.\n- Prefer square or near-square for individual social posts, icons, avatars,\n compact logos, or compact product assets; vertical for covers, posters, phone\n stories, or mobile wallpapers; horizontal for slides, headers, banners, or\n explicitly wide displays. Do not invent a platform or deliverable.\n\n3. Primary composition\n- The main subject, action, or information structure outweighs optional context.\n A generic room, park, forest, sky, or cinematic background must not turn a\n focused subject horizontal.\n- Prefer vertical for dominant upright or full-body subjects, tall structures,\n stacked information, rising or falling motion, upward reach, or vertical\n negative space. Prefer horizontal for genuinely side-by-side structure,\n lateral action, broadly distributed subjects, a requested broad environment,\n horizontal sequences, or lateral negative space. Prefer square or near-square\n for compact, centered, radial, symmetrical, or directionally neutral content.\n\n4. Minimal completion and ratio strength\n- If the input is underspecified, form the smallest natural, content-preserving\n composition needed to choose a ratio. Lock it before adding detailed scenery\n or styling; do not invent a panorama, broad environment, or multi-panel layout\n to justify it, and do not revise it because later expansion adds detail.\n- Words such as \"portrait,\" \"cinematic,\" \"video,\" \"UE5,\" \"Weta,\"\n \"photorealistic,\" \"dramatic,\" and \"wide-angle\" do not determine orientation.\n- Use 5:4 or 4:5 for mild direction, 4:3 or 3:4 for moderate direction, and\n 3:2 or 2:3 for clear photographic direction. Reserve 16:9 and 9:16 for\n strongly directional or mobile-first compositions, and 21:9 for genuine\n banners, panoramas, timelines, or explicitly continuous wide layouts. When\n evidence remains weak, prefer 1:1 or a moderate ratio.\n\nChoose and lock the ratio from the requested canvas, deliverable, and smallest\nnatural primary composition before enriching the caption. Later details added by\nthe caption-expansion rules must not retroactively change the ratio.\n\nUniversal hard rules:\n1. Preserve exact requested text, named entities, counts, formulas, data values,\ncomponent names, style, medium, orientation, and left/right/top/bottom order.\n2. Required content must be visible, not implied. If text or a label is requested,\nplace it on the surface and describe where it sits.\n3. For active bodies, solve scene geometry before pose detail. Preserve any\ndomain-native layout first: a sport court, game grid, path, lane, stair, room,\nor interface should keep its recognizable structure unless the input changes\nit. Do not move the target or redesign the ground merely to obtain a preferred\nbody view. Choose one camera only after the scene layout is fixed, then place\nthe source and target in that camera's image frame using screen-left/screen-\nright, upper/lower image position, and nearer/farther depth. Silently trace the\nsource-to-target vector in this fixed view. Derive the actor's camera-relative\norientation from that vector and state it explicitly as front-facing, back-\nfacing, left-facing profile, right-facing profile, front three-quarter, or rear\nthree-quarter, together with a frame-absolute motion direction such as leftward,\nrightward, toward the camera/foreground, or away from the camera/deeper into the\nscene. Never use only a target-relative phrase such as \"facing square 2\" or\n\"looking toward the door\". The mapping follows the chosen geometry rather than\na fixed default: a target at screen-right implies a right-facing or rightward\nbody; a target at screen-left implies a left-facing or leftward body; a target\nfarther into the scene implies a back-facing or rear-three-quarter body moving\naway; and a target nearer the camera implies a front-facing or front-three-\nquarter body moving toward the viewer. These are coordinate mappings, not\ninstructions to relocate every target. Verify the visible body surfaces before\nwriting: movement away must show the back, back of the head, or rear shoulders\nand must not show a frontal chest or full face; movement toward the camera must\nshow the face and chest; lateral movement must read as the corresponding side\nprofile. If the stated source/target positions and visible body surfaces imply\nopposite directions, revise the body orientation instead of changing the scene\nlayout. Keep gaze, chest, torso lean, leading limb, support foot or hand, and\nmotion consistent with the derived orientation and with one frozen phase. A\nlimb cannot be simultaneously ahead and behind, planted and airborne, or\nassigned to two contacts. For actions on marked ground, use an elevated high-angle oblique or near-\noverhead three-quarter camera, never a low ground-level view. Center the camera on the complete ground map and make it the dominant, fully\ncontained region of the image: show its outer boundary, every cell or segment,\nthe source, target, and full travel path. After deriving the source-to-target\nvector, place the actor's torso outside the map on the ray extending backward\nfrom the source in the direction opposite the target. Do not move the actor to\nan arbitrary side margin if that would change the apparent travel direction.\nOnly the support foot may enter the near edge of the source cell, leaving its\nlabel visible; place the leading foot in the source-target direction near their\ngap. Keep torso, limbs, hair, clothing, and cast shadow outside every later cell\nand label. Widen or elevate the camera framing rather than rotating the actor or\nrelocating the target. This placement must preserve the previously derived front,\nback, left, right, or three-quarter camera-relative orientation. Preserve conventional topology and perspective, keep every label upright and\nlarge enough to read, and show each requested label exactly once in order. If\nthe input does not specify label colors or materials, use one neutral, uniform,\nhigh-contrast marking style for every label; unspecified styling is not permission\nto assign a different color to each item. Keep the actor's hands empty unless the\ninput names a held object. Do not add loose markers, stones, chalk pieces, buckets,\nropes, toys, shadows, limbs, or cropping over any label or grid line.\nPreserve a plausible silhouette and avoid broken foreshortening, hidden critical\njoints, mirrored layouts, alternative camera choices, and blended before-and-\nafter poses. For emotion-driven photo or video frames, keep facial anatomy\nnatural and consistent with the selected viewpoint.\n4. For hands, grips, linked fingers, hugs, carrying, pointing, tools, weapons, or\nbody-object contact, choose the framing that makes the interaction readable. Use\na medium shot or waist-up view when hand/facial detail is central; do not force a\nfull-body view if it makes the required contact or expression tiny. State which\nhand or limb touches what, how the fingers or support points align, and keep\nnonessential limbs simple.\n5. Use world knowledge as visible evidence only: cultural motifs, scientific\nstructures, medical anatomy, interface conventions, manufacturing details,\nprofessional tools, materials, units, labels, measurement marks, clothing, props,\narchitecture, or environment. Do not add lore that cannot be seen.\n6. Respect requested subject count. If the prompt names event types, two\ncharacters, or one primary subject, keep the number of primary figures to that\nminimum unless extra people are requested. Background crowds, spectators, or\nstaff should read as small silhouettes or graphic texture, not extra competing\nmain subjects.\n\nFor communication artifacts, silently make a content map before writing. Divide\nthe canvas in reading order into major slots: header/title, subtitle/tagline,\nhero visual, left panel, right sidebar, card row, icon group, diagram body,\nprocess path, chart area, formula box, exploded view, inset, specification area,\nswatch strip, instruction cards, footer, screen region, or background context.\nEach major slot should normally include at least three of these: position, visible\ntext or label, concrete visual subject, function or relationship, material/data\ntreatment, and hierarchy/style treatment. Do not use vague slot fillers such as\n\"icons,\" \"details,\" \"information,\" \"supporting graphics,\" or \"annotations\"\nwithout saying what they show.\n\nMake the artifact feel complete. If the prompt asks for a card, board, poster,\ninfographic, webpage, or product sheet, add a few plausible supporting elements\nthat belong to that artifact when they help completeness: subtitle, role/profile,\nabout note, quote, ratings, legend, color key, units, variable definitions,\ndimension labels, back/side view, material stack, use-case icons, footer tips,\nQR/logo placeholder, warning, or compact specification table. Keep these as short\nreadable on-image labels or one-line captions, not prose blocks.\n\nCorrespondence and legibility rules:\n- Multi-view outputs such as floor plan plus interior render, blueprint plus\nproduct render, map plus scene, or before/after panels need a shared inventory\nand coordinate frame. Reuse the same walls, openings, furniture/components,\nmaterials, labels, and relative positions in every panel. Name the render camera\norigin or viewing direction from the plan when useful, and do not invent major\nobjects that are absent from the paired view.\n- Dense flowcharts, medical diagrams, route maps, and process infographics should\nbe simplified into a few large numbered nodes, separated lanes, or a clean\nU-shaped loop. Arrows must be thick, directional, non-crossing, and attached to\nnamed start/end nodes. Prefer legible topology and large labels over many small\nleader lines, tiny arrows, or decorative clutter.\n\nContract affordances:\n- Scientific/medical/educational infographic: include named section headers;\nmain diagram with attached component labels; arrows with correct direction and\nstart/end nodes; legends or color keys; short functional captions; key-point\ncards with icon plus meaning; definitions and units; formula box with variable\ndefinitions; and process/system context such as body, lungs, vessels, lab setup,\nmolecule, device, or environment when implied.\n- Charts, graphs, dashboards, and data screens: include chart title when useful,\naxis labels, tick labels or scale endpoints, units, legend entries, colors, data\nmarks, selected states, and values or qualitative endpoints placed consistently\nwith the visual geometry. Even if the chart appears on a photographed stage\nscreen, the screen content still needs readable chart semantics.\n- Product/cultural design board: include title/subtitle/tagline; inspiration or\nuse-case cues; large hero render; motif-specific details; front/back/side or\nscale insets when useful; exploded layers; manufacturable joints, pins, clasps,\nfasteners, seams, ribs, holes, tolerances, or contact points; material stack;\nfinish behavior; dimension labels; palette swatches; specification and craft\nicons.\n- Character card/mascot sheet: include identity/title; name tag; role/profile or\nabout area; full-body readable proportions; attribute rows with icon, label, and\nbrief meaning; props tied to the role; material swatches or callouts; optional\nrating, quote, badge, instruction row, or contextual background so it reads as a\ndesigned card/poster rather than a plain character portrait.\n- UI/web/dashboard/data graphic: include real page/screen sections, navigation,\ncontrols, fields, buttons, cards, selected states, labels, values, icons, and\nintentional empty states only when requested.\n- Public poster/cover/packaging/display graphic: include typography hierarchy,\nheadline/subtitle/supporting copy, image slots, instruction cards, icons, badges,\nborders/rules, QR/barcode/logo placeholders if requested, palette, and intentional\nnegative space. When a poster features sports, products, or people, use one clear\nhero subject per requested category unless the prompt asks for a group.\n\nStyle comes after content. Choose clean isolation for icons/logos/cutouts and\ntechnical assets; layout-rich pages or boards for design communication; believable\nenvironment, lighting, lens, and depth for photographs. Replace abstract mood and\ngeneric adjectives with specific visible cues.\n\nFinal self-check: every input requirement is visible; every major slot has real\ncontent; required text/formulas/values are exact; labels attach to correct\nstructures; arrows and data geometry are coherent; body orientation is explicit relative to the camera and agrees with the preserved scene layout, source, target, depth, gaze, support, and motion; paired panels correspond; subject count has not grown by accident;\nworld knowledge appears as visible evidence; no important surface is blank.\n\nFinal output contract:\nUse the private thinking section to plan the caption. The final canvas aspect ratio is already fixed by the user and is given after the input; compose for it and do not choose another. After thinking, write only the caption: one fluent, detailed English paragraph that follows every rule above. No JSON, no labels, no Markdown, no lists, no alternatives, and no text before or after the caption.\n\ninput_text: {c}\nFinal canvas shape: width divided by height is {a:.3g} (below 1 is portrait, above 1 is landscape)\noutput_text:\n"
3740 },
3741 "color": "#14424d",
3742 "bgcolor": "#161f21"
3743 },
3744 {
3745 "id": 403,
3746 "type": "AUSBOSS_NODES_Callout",
3747 "pos": [
3748 560,
3749 80
3750 ],
3751 "size": [
3752 460,
3753 178
3754 ],
3755 "flags": {
3756 "pinned": true
3757 },
3758 "order": 1,
3759 "mode": 0,
3760 "inputs": [
3761 {
3762 "localized_name": "text",
3763 "name": "text",
3764 "type": "STRING",
3765 "widget": {
3766 "name": "text"
3767 },
3768 "link": null
3769 }
3770 ],
3771 "outputs": [],
3772 "title": "Callout",
3773 "properties": {
3774 "Node name for S&R": "AUSBOSS_NODES_Callout",
3775 "cnr_id": "ausboss-nodes"
3776 },
3777 "widgets_values": [
3778 "For an edit, the writer looks at your picture and any extra pictures, with MiniMax's own edit instructions.\n###### They are counted, and the writer gets one <Picture N> block for each: only as many as you plugged in."
3779 ],
3780 "widgets_values_named": {
3781 "text": "For an edit, the writer looks at your picture and any extra pictures, with MiniMax's own edit instructions.\n###### They are counted, and the writer gets one <Picture N> block for each: only as many as you plugged in."
3782 },
3783 "color": "#07090a",
3784 "bgcolor": "#07090a"
3785 },
3786 {
3787 "id": 404,
3788 "type": "ImageScaleToTotalPixels",
3789 "pos": [
3790 560,
3791 476
3792 ],
3793 "size": [
3794 460,
3795 130
3796 ],
3797 "flags": [],
3798 "order": 6,
3799 "mode": 0,
3800 "inputs": [
3801 {
3802 "localized_name": "image",
3803 "name": "image",
3804 "type": "IMAGE",
3805 "link": 1072
3806 },
3807 {
3808 "localized_name": "upscale_method",
3809 "name": "upscale_method",
3810 "type": "COMBO",
3811 "widget": {
3812 "name": "upscale_method"
3813 },
3814 "link": null
3815 },
3816 {
3817 "localized_name": "megapixels",
3818 "name": "megapixels",
3819 "type": "FLOAT",
3820 "widget": {
3821 "name": "megapixels"
3822 },
3823 "link": null
3824 },
3825 {
3826 "localized_name": "resolution_steps",
3827 "name": "resolution_steps",
3828 "type": "INT",
3829 "widget": {
3830 "name": "resolution_steps"
3831 },
3832 "link": null
3833 }
3834 ],
3835 "outputs": [
3836 {
3837 "localized_name": "IMAGE",
3838 "name": "IMAGE",
3839 "type": "IMAGE",
3840 "links": [
3841 1035
3842 ]
3843 }
3844 ],
3845 "title": "Scale Image - Your picture at 1 MP for the writer",
3846 "properties": {
3847 "Node name for S&R": "ImageScaleToTotalPixels",
3848 "cnr_id": "comfy-core",
3849 "ver": "0.44.0"
3850 },
3851 "widgets_values": [
3852 "lanczos",
3853 1,
3854 1
3855 ],
3856 "widgets_values_named": {
3857 "upscale_method": "lanczos",
3858 "megapixels": 1,
3859 "resolution_steps": 1
3860 },
3861 "color": "#14424d",
3862 "bgcolor": "#161f21"
3863 },
3864 {
3865 "id": 405,
3866 "type": "BatchImagesNode",
3867 "pos": [
3868 560,
3869 676
3870 ],
3871 "size": [
3872 460,
3873 230
3874 ],
3875 "flags": [],
3876 "order": 7,
3877 "mode": 0,
3878 "inputs": [
3879 {
3880 "localized_name": "image0",
3881 "name": "images.image0",
3882 "type": "IMAGE",
3883 "link": 1035
3884 },
3885 {
3886 "localized_name": "image1",
3887 "name": "images.image1",
3888 "shape": 7,
3889 "type": "IMAGE",
3890 "link": 1036
3891 },
3892 {
3893 "localized_name": "image2",
3894 "name": "images.image2",
3895 "shape": 7,
3896 "type": "IMAGE",
3897 "link": 1037
3898 },
3899 {
3900 "localized_name": "image3",
3901 "name": "images.image3",
3902 "shape": 7,
3903 "type": "IMAGE",
3904 "link": 1038
3905 },
3906 {
3907 "localized_name": "image4",
3908 "name": "images.image4",
3909 "shape": 7,
3910 "type": "IMAGE",
3911 "link": 1039
3912 },
3913 {
3914 "localized_name": "image5",
3915 "name": "images.image5",
3916 "shape": 7,
3917 "type": "IMAGE",
3918 "link": 1040
3919 },
3920 {
3921 "localized_name": "image6",
3922 "name": "images.image6",
3923 "shape": 7,
3924 "type": "IMAGE",
3925 "link": 1041
3926 },
3927 {
3928 "localized_name": "image7",
3929 "name": "images.image7",
3930 "shape": 7,
3931 "type": "IMAGE",
3932 "link": 1042
3933 },
3934 {
3935 "localized_name": "image8",
3936 "name": "images.image8",
3937 "shape": 7,
3938 "type": "IMAGE",
3939 "link": null
3940 }
3941 ],
3942 "outputs": [
3943 {
3944 "localized_name": "IMAGE",
3945 "name": "IMAGE",
3946 "type": "IMAGE",
3947 "links": [
3948 1043,
3949 1052
3950 ]
3951 }
3952 ],
3953 "title": "Batch Images - Your picture, then the extra pictures",
3954 "properties": {
3955 "Node name for S&R": "BatchImagesNode",
3956 "cnr_id": "comfy-core",
3957 "ver": "0.44.0"
3958 },
3959 "color": "#14424d",
3960 "bgcolor": "#161f21"
3961 },
3962 {
3963 "id": 406,
3964 "type": "GetImageSize",
3965 "pos": [
3966 560,
3967 976
3968 ],
3969 "size": [
3970 460,
3971 90
3972 ],
3973 "flags": [],
3974 "order": 8,
3975 "mode": 0,
3976 "inputs": [
3977 {
3978 "localized_name": "image",
3979 "name": "image",
3980 "type": "IMAGE",
3981 "link": 1043
3982 }
3983 ],
3984 "outputs": [
3985 {
3986 "localized_name": "width",
3987 "name": "width",
3988 "type": "INT",
3989 "links": []
3990 },
3991 {
3992 "localized_name": "height",
3993 "name": "height",
3994 "type": "INT",
3995 "links": []
3996 },
3997 {
3998 "localized_name": "batch_size",
3999 "name": "batch_size",
4000 "type": "INT",
4001 "links": [
4002 1044,
4003 1141
4004 ]
4005 }
4006 ],
4007 "title": "Get Image Size - How many pictures",
4008 "properties": {
4009 "Node name for S&R": "GetImageSize",
4010 "cnr_id": "comfy-core",
4011 "ver": "0.44.0"
4012 },
4013 "color": "#14424d",
4014 "bgcolor": "#161f21"
4015 },
4016 {
4017 "id": 407,
4018 "type": "AUSBOSS_NODES_MathExpression",
4019 "pos": [
4020 560,
4021 1136
4022 ],
4023 "size": [
4024 460,
4025 134
4026 ],
4027 "flags": [],
4028 "order": 9,
4029 "mode": 0,
4030 "inputs": [
4031 {
4032 "localized_name": "a",
4033 "name": "a",
4034 "shape": 7,
4035 "type": "FLOAT,INT",
4036 "link": 1044
4037 },
4038 {
4039 "localized_name": "b",
4040 "name": "b",
4041 "shape": 7,
4042 "type": "FLOAT,INT",
4043 "link": null
4044 },
4045 {
4046 "localized_name": "c",
4047 "name": "c",
4048 "shape": 7,
4049 "type": "FLOAT,INT",
4050 "link": null
4051 },
4052 {
4053 "localized_name": "expression",
4054 "name": "expression",
4055 "type": "STRING",
4056 "widget": {
4057 "name": "expression"
4058 },
4059 "link": null
4060 }
4061 ],
4062 "outputs": [
4063 {
4064 "localized_name": "float",
4065 "name": "float",
4066 "type": "FLOAT",
4067 "links": []
4068 },
4069 {
4070 "localized_name": "int",
4071 "name": "int",
4072 "type": "INT",
4073 "links": [
4074 1046
4075 ]
4076 }
4077 ],
4078 "title": "Math Expression - Letters for that many blocks",
4079 "properties": {
4080 "Node name for S&R": "AUSBOSS_NODES_MathExpression",
4081 "cnr_id": "ausboss-nodes",
4082 "ver": "2.15.0"
4083 },
4084 "widgets_values": [
4085 "a * 102"
4086 ],
4087 "widgets_values_named": {
4088 "expression": "a * 102"
4089 },
4090 "color": "#14424d",
4091 "bgcolor": "#161f21"
4092 },
4093 {
4094 "id": 408,
4095 "type": "AUSBOSS_NODES_Text",
4096 "pos": [
4097 560,
4098 1340
4099 ],
4100 "size": [
4101 460,
4102 168
4103 ],
4104 "flags": [],
4105 "order": 2,
4106 "mode": 0,
4107 "inputs": [
4108 {
4109 "localized_name": "text",
4110 "name": "text",
4111 "type": "STRING",
4112 "widget": {
4113 "name": "text"
4114 },
4115 "link": null
4116 }
4117 ],
4118 "outputs": [
4119 {
4120 "localized_name": "text",
4121 "name": "text",
4122 "type": "STRING",
4123 "links": [
4124 1045
4125 ]
4126 }
4127 ],
4128 "title": "Text - One block per picture, cut to the count",
4129 "properties": {
4130 "Node name for S&R": "AUSBOSS_NODES_Text",
4131 "cnr_id": "ausboss-nodes",
4132 "ver": "2.15.0"
4133 },
4134 "widgets_values": [
4135 "Reference Media:\n- <Picture 1> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 2> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 3> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 4> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 5> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 6> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 7> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 8> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\n"
4136 ],
4137 "widgets_values_named": {
4138 "text": "Reference Media:\n- <Picture 1> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 2> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 3> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 4> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 5> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 6> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 7> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\nReference Media:\n- <Picture 8> [aspect_ratio unavailable]\n<|vision_start|><|image_pad|><|vision_end|>\n"
4139 },
4140 "color": "#14424d",
4141 "bgcolor": "#161f21"
4142 },
4143 {
4144 "id": 409,
4145 "type": "StringFormat",
4146 "pos": [
4147 560,
4148 1570
4149 ],
4150 "size": [
4151 460,
4152 190
4153 ],
4154 "flags": [],
4155 "order": 10,
4156 "mode": 0,
4157 "inputs": [
4158 {
4159 "localized_name": "a",
4160 "name": "values.a",
4161 "shape": 7,
4162 "type": "*",
4163 "link": 1045
4164 },
4165 {
4166 "localized_name": "b",
4167 "name": "values.b",
4168 "shape": 7,
4169 "type": "*",
4170 "link": 1046
4171 },
4172 {
4173 "localized_name": "f_string",
4174 "name": "f_string",
4175 "type": "STRING",
4176 "widget": {
4177 "name": "f_string"
4178 },
4179 "link": null
4180 },
4181 {
4182 "localized_name": "c",
4183 "name": "values.c",
4184 "shape": 7,
4185 "type": "*",
4186 "link": 1047
4187 },
4188 {
4189 "localized_name": "d",
4190 "name": "values.d",
4191 "shape": 7,
4192 "type": "*",
4193 "link": 1141
4194 }
4195 ],
4196 "outputs": [
4197 {
4198 "localized_name": "STRING",
4199 "name": "STRING",
4200 "type": "STRING",
4201 "links": [
4202 1051
4203 ]
4204 }
4205 ],
4206 "title": "Format Text - MiniMax's edit instructions, the pictures, your words",
4207 "properties": {
4208 "Node name for S&R": "StringFormat",
4209 "cnr_id": "comfy-core",
4210 "ver": "0.44.0"
4211 },
4212 "widgets_values": [
4213 "<|im_start|>user\nYou are an expert at enhancing a short image-editing instruction into a rich, vivid, reference-conditioned description of the intended finished target image.\n\n## What you are given\n- Input image(s), labeled sequentially starting from <Picture 1>. These are the user's source and reference images.\n- One short editing instruction. It may be terse, vague, informal, or phrased as a desired result rather than an imperative. The final text part supplies it as User Editing Instruction (JSON string).\n- User Aspect Ratio Selection: any concrete positive-integer W:H ratio, or adaptive, supplied in the final text part.\n- Each input image is immediately preceded by a Reference Media text part containing its <Picture N> label and aspect_ratio W:H. The caller has already reduced the image's original display width and height to their exact lowest integer ratio, after EXIF orientation and before preview resizing or padding. No pixel dimensions are supplied in this metadata. The ratio applies only to the immediately following image. The supplied aspect_ratio is the reference image's exact ratio, not necessarily the final target ratio. Use it as given; do not estimate the source ratio from the preview. Determine the final target ratio using the rules below. If the original dimensions are unknown, the metadata says aspect_ratio unavailable.\n- There is no target image. Construct the intended target from the visible source evidence and the instruction; never pretend to have observed a hidden result.\n\n## Response format\nReturn exactly one valid JSON object with exactly these two keys, both with string values:\n{{\"expanded_caption\":\"COMPLETE_DESCRIPTION\",\"target_aspect_ratio\":\"W:H\"}}\nResolve target_aspect_ratio using the rules below. Put the complete description in expanded_caption as one continuous paragraph with no actual line breaks. All narration, language, Subject, literal-text, and three-part rules below apply to expanded_caption, not to the outer JSON syntax or target_aspect_ratio. Use valid JSON escaping for quotes and backslashes without changing the decoded literal content. Do not output reasoning, a code fence, extra keys, or text outside this JSON object.\n\n## Target aspect ratio\nDetermine target_aspect_ratio before writing expanded_caption. Treat it as a constraint on the intended target image, adapting framing and composition while respecting the requested edit and reference-preservation requirements. Describe any necessary changes under the changed/new-content rules. User Aspect Ratio Selection is part of the user's request. A concrete selection governs conflicting canvas wording in the editing instruction; continue to honor all compatible content constraints. Do not mechanically copy reference metadata or routing explanations into expanded_caption, and do not treat metadata as lettering inside an image.\n\nReading aspect_ratio metadata alone must not create a <Subject N> or change a preservation relationship. However, when source framing, aspect ratio, or composition is independently preserved, changed, or borrowed, identify that source-derived scope under the rules below. If realizing the selected canvas actually changes framing, visible content, placement, or layout within a Subject's defined scope, reflect that change in its relationship and describe only the changed/new consequences in Part 3. Do not call the entire source composition fully preserved while changing it; do not automatically mark every Subject partially preserved merely because the canvas ratio differs. Limit any completion or reframing to what the user's edit and selected canvas require; preserve unrelated content.\n\nRead User Aspect Ratio Selection from the final input part.\n\nSpecified ratio:\nIf a concrete ratio is supplied, copy that exact string to target_aspect_ratio. Any positive-integer W:H ratio is allowed; it is not limited to a preset list. Neither the editing instruction nor a reference image may override it.\n\nAdaptive mode:\nOnly when the selection is adaptive, determine the aspect ratio of the intended finished target image, which may differ from the main reference image. Apply these rules in order:\n\n1. Follow an explicit final target ratio or target dimensions in the editing instruction. Use the requested final geometry rather than source or earlier geometry. Exact numeric geometry takes precedence over generic orientation. When only a final orientation is specified, use 21:9 for ultra-wide, 16:9 for landscape, 1:1 for square, and 9:16 for portrait.\n\n2. Otherwise, select the main reference that supplies the edited base canvas or overall target composition and has a usable aspect_ratio. Prioritize an explicitly designated canvas/composition reference, then the edited base, then an overall composition anchor. Break ties by Picture order and skip missing, invalid, or unavailable ratios. Images used only for identity, objects, color, or style do not determine the canvas.\n\n Use the main reference's aspect_ratio as the starting ratio. Copy it exactly unless the requested edit requires a different canvas shape. Proportional resizing or changing a subject's size alone does not require changing the ratio.\n\n When expansion or cropping changes the canvas shape, determine the ratio of the resulting target. If the available information unambiguously determines it, calculate it exactly and reduce it. For relative changes:\n target_ratio = reference_ratio × width_scale / height_scale,\n where both scales are relative to the original canvas.\n\n If the requested shape change does not determine a unique ratio, choose one that fits the intended result while preserving unrelated composition. Adjust it only as needed for the requested edit, not merely for aesthetics.\n\n3. If neither an explicit target requirement nor a usable main reference determines the ratio, output 16:9.\n\nApply these rules to the whole target canvas. Grid row/column counts, individual panels, objects inside the image, and literal text are not whole-canvas ratio requests. A portrait subject or wide-angle lens alone is not a target-orientation request. Do not use a reference board's ratio as the ratio of one panel.\n\nSerialize target_aspect_ratio as two positive integers in W:H form, with no spaces; never output adaptive. Preserve concrete user selections, explicit positive-integer target ratios in the instruction, and unchanged reference ratios exactly. Convert other explicit numeric ratios or target dimensions to their exact reduced integer ratio; reduce newly calculated or chosen ratios to their lowest integer form. Do not snap to preset ratios or clamp to a supported range.\n\nLater descriptive additions must not change the chosen ratio. Pixel-size resolution, 32 alignment, and downstream size fallback are outside this task.\n\n## Interpret the User's Intent Flexibly\nRead the instruction together with the input images and any context it supplies to understand the result the user actually wants. Terse wording, colloquial phrasing, shorthand, and requests such as 'complete this' may require reasonable interpretation and substantive completion. Do not mechanically echo the request, insist that the user prewrite every detail, or treat an illustrative example or category list as the only supported case.\n\nChoose the interpretation best supported by the visible evidence, the request's purpose, and ordinary context. Supply the visual or textual details needed to make that intended result complete and coherent, including directly supporting consequences. Be flexible in how you realize the request, but do not override explicit constraints, change specified literal wording without permission, invent source facts, or expand unrelated content. Keep the boundaries between preserved source content, requested changes, newly authored text, and actual image lettering clear.\n\n## Source-Grounded Content and New Target Content\nTreat source-derived content and instruction-introduced content differently. Source facts describe what is actually visible in the inputs; target changes describe the requested result. Source fidelity does not prohibit an explicitly requested transformation. Preservation is the default only outside requested changes and their necessary consequences.\n\nA) Descriptions of source facts and properties retained in the target must be strictly faithful. Ground identity, facial features, body shape, clothing, objects, colors, materials, text, data, count, pose, layout, scene, camera, lighting, and spatial relationships in the specific input picture where they are visible. Do not invent a source attribute, accessory, logo, word, number, object, or relationship. If something is genuinely illegible or ambiguous, describe that fact neutrally rather than guessing. When an unfamiliar source element lacks a reliable name, identify it using its location and distinctive visible cues rather than guessing its identity or function.\n\nB) Content newly introduced or substantially transformed by the instruction should still be imagined richly, concretely, and coherently. Within the changed or new scope, supply plausible visual specifics that make the requested result recognizable and renderable: materials, shapes, arrangement, depth, perspective, lighting direction, shadows, reflections, contact, overlap, occlusion, atmosphere, and surrounding details that naturally belong to the requested scene. A newly requested background, environment, design, or global style may be described as a complete vivid visual scope because that scope itself is new or changed. Do not make the answer skeletal merely to satisfy a delta format.\n\nWhenever the intended target needs visual content not already determined by a preserved Subject or an explicitly referenced property, reasonably imagine and concretely describe that new content. This applies to any new scene, entity, action, design, composition, or newly visible region, not only viewpoint changes. Treat unspecified details within that new scope as room for coherent creation, not a reason to leave it vague. Keep the invention consistent with the user's intent and remaining reference constraints; do not redesign preserved Subjects or alter properties that the request requires you to match. If a source Subject gains newly visible parts, keep its identity and Subject number; describe the plausible completion in Part 3, never as an observed source fact.\n\nCreative latitude stays inside the requested or directly supporting scope. Do not redesign unrelated source content or import incidental attributes from a donor picture merely to enrich the prose. Ground factual claims in the available reference and instruction; do not invent unsupported specifics and present them as real. Explicitly requested fictional names or illustrative sample data may be authored as fictional or illustrative target content, never as recovered source facts or verified real-world information. Omit unsupported specifics, not the requested writing itself, and do not add unrelated text beyond the requested scope.\n\nComplete every textual deliverable the user asks you to write, supplement, finish, expand, rewrite, or otherwise provide. If exact wording is not supplied, compose the complete final wording yourself; missing wording is not a reason to leave the content unspecified. This applies to any requested textual deliverable, not limited to particular genres or to text inside an image. Requested wording is the deliverable itself, not forbidden restatement of the user's instruction. Do not delegate completion to the downstream model.\n\n## Output contract\nWrite expanded_caption as one continuous plain-text paragraph. Do not add structural headings, labels, bullets, markdown, preamble, reasoning, notes, or meta-commentary to the narration. Literal content may contain such words or symbols when they belong to source text or requested writing. The paragraph has three logical parts, in this order:\n1. natural definitions of the relevant source-derived <Subject N> elements;\n2. one target overview followed by a separate preservation/change relationship clause for every defined Subject;\n3. Changed and Newly Introduced Target Details Only, together with the complete wording of any textual deliverable the user asks you to provide.\n\nDetermine each requested text's destination from the instruction. If it belongs in the image, include its literal wording and describe placement as needed. If it is requested only as a textual response, provide it as text alongside the visual delta; do not turn it into a sign, caption, label, or other image element. Quoting a textual deliverable does not mean that it should be rendered in the image.\n\nComplete all Subject definitions before the target overview. Do not expose structural commands such as 'Define', 'Let', 'Identify', 'Section 1', 'Section 2', or 'Section 3' as narration. Write natural element-first definitions, for example, 'The woman in a red coat in <Picture 1> is <Subject 1>.' Refer to the intended visual result only as 'the target image'. Narrate the result, not the user's instruction or an imperative such as 'generate' or 'please create'; this does not prohibit exact supplied text or requested authored text from containing those words. If no source-derived element is retained or referenced, omit Subject definitions and relationships and begin with the target overview; do not invent a Subject to fill the structure.\n\n## Part 1: Natural Reference Element Identification\nAssign <Subject N> only to source-derived elements that remain visible or traceable in the intended target, or that provide a visible transferred or borrowed property. Every definition must name its source with an explicit <Picture N> and concise visible cues sufficient to uniquely identify the referenced object, region, or property. Add detail only to resolve ambiguity or clarify the reference/edit scope, not to restate its full appearance. Keep the full text/data transcription required below; concise identification must not omit readable content. List source pictures individually; never use ranges or vague phrases such as 'the reference images', 'the previous images', 'all inputs', or 'these pictures'. Use one number for the same convincingly identified person or element across pictures, and keep clearly different identities separate.\n\nWhen only a source property is needed, define that property rather than the entire source object; use the source object or region to locate the property. Give the source object its own Subject only if it is independently used in the target.\n\nChoose Subject granularity by relationship scope, not by a quota. An element that is independently moved, recolored, restyled, reposed, recounted, replaced while remaining traceable, or used as a donor should be independently addressable. Split a part, pose, layout, palette, lighting setup, text block, garment, accessory, held object, or background only when it needs its own preservation/change judgment. Do not mechanically split an ordinary person into face, hair, hands, every garment, and every accessory when those parts move together and have no independent role. Do not force one Subject per picture, and do not define incidental objects merely to make the inventory longer.\n\nJudge preservation over the scope actually defined. A parent Subject cannot be fully preserved if a changed child remains inside its scope, even when the child has a separate number. A genuinely unchanged narrower region or identity may be defined explicitly, but do not call the whole person, object, or scene unchanged while editing part of it. Do not transcribe the same source field again through an overlapping parent or child Subject.\n\nThe identifiable types include the following when independently relevant:\n- Entity type: a person, animal, object, creature, garment, accessory, prop, vehicle, furniture item, or other concrete entity.\n- Scene/background type: a room, landscape, street, stage, tabletop, architecture, or other environment preserved, changed, or used by the target.\n- Attribute/style type: a source-derived texture, material language, visual medium, design style, typography, border treatment, or other transferable visible quality.\n- Shared-element type: one convincingly identical visual element shown across several pictures; define it once and list every source picture.\n- Multi-form reference type: one source element used in several intended target forms, such as a physical figure, package artwork, screen render, or line drawing; define it once and distinguish the forms later.\n- Data-bearing chart/table/plot/document type: do not collapse readable data into one undifferentiated Subject. Identify the source-derived series, bars, lines, slices, columns, labeled groups, axes, titles, legends, headers, fields, and readable cells needed to preserve or edit the visible data faithfully.\n- Dense/multi-line text type: identify every clearly readable source-derived line or field in a document, slide, poster, menu, search interface, caption block, subtitle group, or UI region. Either define lines separately or define one coherent block while transcribing its lines literally and in order; do not replace readable content with a vague 'text' Subject.\n- Text/watermark type: identify an isolated watermark, brand text, subtitle, label, sign, logo text, or other short readable element, and state its clearly readable literal content once in Part 1. Later use its Subject token unless that text is intended to change.\n- Part/region type: hair, a sleeve, a wheel, a screen area, a poster corner, a body region, a panel, or another local part that needs its own relationship judgment.\n- Spatial-layout/composition type: framing, aspect ratio, grid or split-screen geometry, panel order, placement scheme, image-text hierarchy, or whitespace when the arrangement itself is preserved, changed, or borrowed.\n- Color/palette/lighting type: a source color, palette, color grade, illumination direction, warmth, contrast, shadow design, or lighting setup when color or light itself is independently referenced. Locate the specific source object or region; for example, 'The color of the cup's body in <Picture 1> is <Subject 1>.'\n- Pose/action type: a source-derived posture, gesture, gaze, body orientation, action, or interaction that is preserved, changed, or transferred.\n- Logo/symbol/icon type: a graphical emblem, symbol, UI icon, or pictographic mark; handle any readable words under the text rules.\n- Embedded/thumbnail type: a complete or cropped source picture embedded in the target as a panel or thumbnail.\n- Partial-crop type: a target crop or zoom whose visible content comes from a specific region of a source picture.\n\nThese fifteen types form the complete supported taxonomy, not a per-example checklist or quota. Use every applicable type supported by the input images and the instruction when it needs an independent relationship judgment, but never create an inapplicable Subject merely to instantiate a type. The more specific part/layout/light/pose/icon types exist to prevent an independently changed unit from being buried inside an oversized Subject, not to force mechanical over-splitting.\n\nBrand-new target-only content has no <Subject N>. A source element explicitly removed or wholly replaced and not otherwise reused also has no Subject number; if necessary, identify it once with its <Picture N> in the target overview, then describe the positive completed region in Part 3. Editing or translating a source text field is a traceable change to that field, not removal of the entire element: retain its Subject when needed to locate the edit, record its original wording in Part 1, and give its full resulting wording in Part 3. Every defined Subject must receive a relationship later and must not be defined merely to say that it is absent, removed, unused, or forgotten.\n\nReadable source text and data require special care. For every defined source-derived chart, table, document, UI, text block, or isolated text element, transcribe exactly once in Part 1 every clearly readable source-state line, character, label, value, axis title, axis tick, legend entry, table header, field, and readable cell, including a source literal that the instruction intends to change. Preserve the original language, script, grouping, and reading order. Do not omit a clearly readable value merely because it is unchanged or peripheral to the main edit. Never guess hidden, cropped-away, or illegible source text.\n\nFor grouped or clustered bars, radar charts, multi-panel plots, and many-point plots, enumerate every clearly readable value rather than summarizing the display. Check counts only when the chart's visible semantics require them: grouped bars against the legend, radar values against the axes, and mutually exclusive percentage parts of a complete whole against approximately 100 percent. Do not impose those assumptions on unrelated chart types or incomplete, overlapping, or multi-select categories. If the source itself contains a genuine inconsistency, missing series, or illegible value, record only the visible evidence or the uncertainty in Part 1; never silently normalize or repair the source record. An explicitly requested correction belongs in the target changes and must use supplied or derivable values, not guessed real data. Exhaustive transcription does not require one Subject per value or line: a coherent series, row, field group, or text block may remain one Subject when its literals are fully enumerated and it does not need separate relationship judgments. These checks apply only where the relevant marks and values are actually readable, and they do not require turning an ordinary non-data image into a chart inventory.\n\n## Part 2: Target Overview and Reference Relationships\nAfter the definitions, write one natural sentence beginning with 'The target image'. Foreground the primary intended changed or new state and describe the overall result using the defined Subject tokens where appropriate. Target-only new content remains unnumbered. The overview may summarize several requested changes, but it should remain an overview rather than a second detailed caption.\n\nThen give every defined Subject exactly one whole-subject relationship in its own independent clause. Give each Subject its own relationship clause; do not combine several Subjects into one relationship statement, even when they share the same degree. Keep the prose natural without imposing a fixed sentence length, a fixed Subject count, or a target frequency for any relationship type. Use these four relationships:\n\n- Fully preserved: use only when the whole Subject is unchanged in identity, appearance, count, pose or state, position, topology, and intended target role. Write only the short standalone clause '<Subject N> is fully preserved.' Give each fully-preserved Subject its own sentence; never merge multiple Subjects into one fully-preserved clause. Do not append a checklist, explanation, source citation, or visual description to a fully-preserved clause.\n- Partially preserved: use when the Subject remains traceable but at least one relevant property changes. State the retained categories needed to anchor identity and the changed categories, including a concrete before-to-after relationship when the source and instruction determine it. Do not exhaustively re-render retained appearance.\n- Attribute transfer: use when a specific source-derived property is directly reused or matched on a target carrier. Name the source Subject, the transferred property, and the target carrier. Transfer only the assigned property, not unrelated donor attributes.\n- Weak reference: use when the source guides a general style, palette, atmosphere, composition principle, or similar quality without requiring a specific property to be matched exactly. Name the borrowed quality and make clear what the new design may vary. Choose by the requested use of the reference, not merely by whether the property is called a color, style, or palette.\n\nAttribute-transfer example: the user wants the coat in <Picture 2> to match the cup's body color in <Picture 1>. Part 1 defines that color as <Subject 1> and the coat as <Subject 2>. The relationship clauses can be '<Subject 1> supplies the exact color transferred to <Subject 2>. <Subject 2> is partially preserved; its color changes while its cut and material remain.' The cup's shape and material are not transferred.\n\nWeak-reference example: the user wants a new abstract composition inspired by a landscape's palette, without copying it exactly. Part 1 defines the landscape's palette as <Subject 1>. The relationship clause can be '<Subject 1> serves as a weak palette reference for the new abstract composition; related colors are redistributed among geometric blocks, with freely adapted proportions and placement.' The new composition has no Subject number. These examples illustrate the distinction, not a restriction of either relationship to color tasks.\n\nFor a Subject used in several target forms or instances, judge the intended uses together. A mix of unchanged and transformed or additional instances is partially preserved overall; distinguish those uses within that Subject's own relationship statement. Do not call the entire Subject fully preserved just because one instance is unchanged. If only a property is borrowed, use the appropriate transfer or weak-reference relationship. If a donor object remains unchanged while its color or another property is transferred elsewhere, define that property separately so the object's preservation and the property's transfer have distinct scopes.\n\nThe relationship must agree with the described target. If a requested change or its necessary integration changes a Subject's position, orientation, count, topology, role, pose, expression, clothing, color, material, lighting, state, or other property within its defined scope, the whole Subject is not fully preserved. If later changed-detail prose genuinely needs to describe a change to a Subject, classify that Subject as partially preserved instead. A source carrier may remain fully preserved while one of its properties is transferred outward; do not mark the carrier partial unless the carrier itself changes. Conversely, new shadows, reflections, occlusion, or relighting on the carrier count as changes where they affect its defined appearance; do not hide them behind a fully-preserved judgment.\n\nFor partially preserved chart, table, document, UI, and text-block Subjects, state which categories are retained and which change. For a fully preserved chart or text Subject, use only the standard fully-preserved sentence, not an extra checklist. Keep relationship statements compact: use Part 1 as the source-side record and leave complete changed or newly authored wording and values for Part 3. Do not duplicate old literals in a before-to-after phrase or new literals in both the relationship and detail blocks.\n\nTreat unmentioned source-derived content as stable by default, but do not let that conservative baseline suppress the requested imagination. The changed/new target scope may still be rich. Direct semantic, spatial, physical, compositional, environmental, camera, and lighting consequences of the requested change are part of the changed scope and may be described.\n\n## Part 3: Changed and Newly Introduced Target Details Only\nThis final logical block is a richly rendered visual delta, not a standalone T2I re-caption of the whole target image. Also supply the requested textual deliverables. For visual description, cover only:\n- the concrete final appearance of source-derived attributes that actually change;\n- target-only people, objects, text regions, environments, panels, charts, layouts, effects, or other content introduced by the instruction;\n- the newly exposed or positively completed surface or space left by a removal, or the positive new occupant of a replacement;\n- plausible supporting details inside the changed/new scope that make it visually recognizable and coherent;\n- direct integration consequences involving scale, orientation, contact, overlap, occlusion, perspective, depth, shadow, reflection, lighting, material response, anatomy, support, camera framing, or composition.\n\nThis changed/new-only rule does not impose terse writing. A newly introduced restaurant, fantasy landscape, product design, comic panel, interface, lighting scheme, or global style may be described vividly and comprehensively because that entire requested scope is new or changed. Likewise, a transformed garment, pose, object, chart, or layout may receive enough concrete detail to be rendered reliably. Be imaginative where the instruction creates room for imagination; simply keep that imagination inside the changed/new scope instead of using it to re-caption preserved source content.\n\nStrict non-repetition rules:\n- After a Subject is classified as fully preserved, never visually describe that Subject again. A later sentence may use only the bare <Subject N> token as the minimum spatial or interaction anchor needed to locate changed or new content, for example, 'A new luminous pearl rests beside <Subject 2>.' Do not attach an adjective, color, material, shape, part, pose, position, role, or other appearance detail to that preserved Subject.\n- For a partially preserved Subject, Part 3 describes only its changed attributes and their concrete intended final appearance. Do not repeat the detailed appearance of retained identity, clothing, body, object, scene, or layout attributes already grounded in Part 1 or named as retained in Part 2.\n- For attribute transfer or weak reference, describe only how the transferred or borrowed quality is realized on the intended target carrier. Do not import or re-render unrelated donor properties.\n- Do not repeat fully preserved source text or data as visual description; use the Part 1 record and bare <Subject N> references. If the user separately requests a complete transcription or quotation as a textual deliverable, write that deliverable in full even when it repeats source wording; this is required output content, not redundant visual description.\n- Write every changed or target-only text/data field and requested textual deliverable in full. Use exact supplied or deterministically derived results; preserve every supplied character within each line, including punctuation, capitalization, order, and language, unless the user explicitly requests translation or another textual transformation, in which case write its actual result. If a line is edited, write the complete resulting line, not merely the changed word or numeric difference; do not recopy separate unaffected lines as visual description. Never substitute a <Subject N> token, a description or writing instruction, placeholders, omitted middles, 'etc.', 'and so on', or ellipses for requested content; preserve an ellipsis that belongs to supplied literal text.\n- Do not describe an unchanged background, person, object, pose, camera, light, style, or image-quality property merely to make the paragraph resemble a complete T2I caption.\n- If no changed or newly introduced detail remains beyond the overview and relationship statements and no requested textual deliverable remains to be supplied, stop. Never pad the output with preserved appearance.\n\nFor example, for a reference showing a lakeside path and hills, a request to write a scenic introduction requires actual copy such as '沿着湖畔步道缓缓前行，欣赏湖水与远山相映的景色，在悠闲漫步中感受旅途的宁静。', not 'a paragraph introducing the attraction'. This illustrates completion and is not a restriction to travel copy or permission to reuse these words in unrelated tasks.\n\nCorrect pattern: 'The black horse in <Picture 1> is <Subject 1>. The target image shows <Subject 1> beside a newly added luminous pearl. <Subject 1> is fully preserved. The pearl emits a cool glow onto the ground beside <Subject 1>.' The final sentence describes the new pearl and its integration while using the bare Subject only as an anchor.\n\nIncorrect pattern: '<Subject 1> is fully preserved. <Subject 1> is a muscular black horse with segmented dark armor and a long mane.' The second sentence is a T2I-style re-caption of preserved content and must be omitted.\n\nCorrect partial pattern: '<Subject 1> is partially preserved; its identity and body armor remain, while its bare head changes to an ornate helmet. The new helmet has a pointed silver crest, a red plume, and shadows that follow the existing head orientation.' Do not subsequently re-describe the retained face, body, or armor.\n\nFor exact reference-color or palette matching, use the defined color or palette <Subject N> to specify where it is applied on the target. The visible reference, not an approximate color name, determines the color; do not replace it with inferred color names or guessed RGB/HEX values. Honor explicit user-supplied colors or values, and do not transfer unrelated donor material or lighting unless requested.\n\nWhen describing a grid, multi-panel image, storyboard, table-like arrangement, or other row-and-column layout, state the row count and column count explicitly, either in words such as '2 rows and 3 columns' or with labeled multiplication notation such as '2 x 3 (rows x columns)' or '3 x 2 (columns x rows)'. Multiplication notation is allowed; specify which number denotes rows and which denotes columns. Do not use an unlabeled '2 x 3' or '2×3' alone. When relevant, describe the cell contents in top-to-bottom row order and left-to-right column order, and name which row and column contains each changed or newly introduced element.\n\nFor uneven rows, merged cells, or a freeform composition, describe the actual arrangement directly, such as 'the top row has three panels and the bottom row has two', with positions and spans as needed; do not impose a rectangular grid that is not intended. If shorthand in the instruction is ambiguous, use the reference and context to interpret the intended arrangement and express the chosen row/column meaning explicitly. Record relevant unchanged source layout in the Subject definitions without repeating it after a fully-preserved judgment; Part 3 describes changed or new layout only. This wording rule does not authorize altering supplied literal text containing dimension notation.\n\n## Final language and self-check\nWrite all descriptive and relationship prose in fluent English, regardless of the language of the user instruction. For newly authored requested copy, use the user's requested language, or otherwise the language of the user's request. Keep literal text and completed requested textual deliverables in single quotes to distinguish them from the surrounding English visual description and relationship prose, preserving apostrophes and other punctuation inside the literal text. Represent multi-line literal content inline as separate quoted lines in reading order, describing the intended line breaks or layout in prose rather than inserting actual newline characters into the response.\n\nUse objective, concrete visual language. Richness should come from precise changed/new appearance and coherent integration, not from repetition. Describe visible expressions rather than asserting unobservable thoughts or feelings. Preserve expressions outside the changed scope; expressions for newly introduced or transformed people may be reasonably imagined within the request. Write reference markers <Picture N> and <Subject N> directly without backticks; do not rewrite a literal text string merely because it contains similar words or symbols.\n\nBefore answering, verify:\n- every source visual fact is visible in a named <Picture N>, and every intended change is supported by the editing instruction or User Aspect Ratio Selection and their necessary consequences; reference aspect ratios come only from the supplied metadata;\n- the output still imagines instruction-introduced content richly enough to be recognizable and renderable;\n- every Subject definition is concise but unambiguous about its source and reference scope, without unnecessary appearance detail or missing required text/data;\n- every Subject is naturally defined before use and has its own independent relationship clause, never merged with another Subject;\n- every fully preserved Subject has its own exact '<Subject N> is fully preserved.' sentence and receives no later visual re-description;\n- no fully preserved parent contains an edited child, and mixed target forms or integration effects are reflected in the relationship;\n- every partially preserved Subject repeats no detailed retained appearance in Part 3;\n- donor attributes do not leak beyond the property requested;\n- target-only and removed-only content has no Subject number;\n- every applicable reference-element type is supported without treating the taxonomy as a per-example quota;\n- source text/data is recorded once in Part 1 without guessing, repairing, or translating the source record;\n- all requested target text and textual deliverables are complete and literal, with source repetition only for an explicitly requested transcription or quotation;\n- each textual deliverable has the requested image or text-only destination;\n- Part 3's visual description contains changed/new content and direct integration only, with no whole-frame T2I padding;\n- all non-literal descriptive and relationship prose is English, while quoted source, user-specified, or newly authored requested text uses its required language and script;\n- any grid layout described makes the row count and column count unambiguous, either in words or with explicitly labeled multiplication notation, with positions or uneven-row structure clarified where needed;\n- expanded_caption is one fluent, complete paragraph with no structural meta-language in the narration; multi-line literals retain their wording, reading order, and intended line boundaries through inline quoted lines;\n- target_aspect_ratio copies any concrete user selection exactly; in adaptive mode, it preserves the main reference ratio when no canvas-shape change is required and reflects any requested expansion or cropping; expanded_caption and its preservation relationships agree with the final canvas;\n- the final answer is a valid JSON object containing only the two string fields target_aspect_ratio and expanded_caption.\n\nOutput only the required two-field JSON object.\n\nReference Image Count:\n{d} input image(s), labeled <Picture 1> through <Picture {d}>.\n\n{a:.{b}}\nUser Editing Instruction (JSON string):\n\"{c}\"\n\nUser Aspect Ratio Selection:\nadaptive\n\nApply the instructions above to the supplied inputs and reference images. Output override: the canvas is already fixed. Write only the expanded_caption text itself: one fluent English paragraph. No JSON, no keys, no braces, no Markdown, nothing before or after the paragraph.<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n"
4214 ],
4215 "widgets_values_named": {
4216 "f_string": "<|im_start|>user\nYou are an expert at enhancing a short image-editing instruction into a rich, vivid, reference-conditioned description of the intended finished target image.\n\n## What you are given\n- Input image(s), labeled sequentially starting from <Picture 1>. These are the user's source and reference images.\n- One short editing instruction. It may be terse, vague, informal, or phrased as a desired result rather than an imperative. The final text part supplies it as User Editing Instruction (JSON string).\n- User Aspect Ratio Selection: any concrete positive-integer W:H ratio, or adaptive, supplied in the final text part.\n- Each input image is immediately preceded by a Reference Media text part containing its <Picture N> label and aspect_ratio W:H. The caller has already reduced the image's original display width and height to their exact lowest integer ratio, after EXIF orientation and before preview resizing or padding. No pixel dimensions are supplied in this metadata. The ratio applies only to the immediately following image. The supplied aspect_ratio is the reference image's exact ratio, not necessarily the final target ratio. Use it as given; do not estimate the source ratio from the preview. Determine the final target ratio using the rules below. If the original dimensions are unknown, the metadata says aspect_ratio unavailable.\n- There is no target image. Construct the intended target from the visible source evidence and the instruction; never pretend to have observed a hidden result.\n\n## Response format\nReturn exactly one valid JSON object with exactly these two keys, both with string values:\n{{\"expanded_caption\":\"COMPLETE_DESCRIPTION\",\"target_aspect_ratio\":\"W:H\"}}\nResolve target_aspect_ratio using the rules below. Put the complete description in expanded_caption as one continuous paragraph with no actual line breaks. All narration, language, Subject, literal-text, and three-part rules below apply to expanded_caption, not to the outer JSON syntax or target_aspect_ratio. Use valid JSON escaping for quotes and backslashes without changing the decoded literal content. Do not output reasoning, a code fence, extra keys, or text outside this JSON object.\n\n## Target aspect ratio\nDetermine target_aspect_ratio before writing expanded_caption. Treat it as a constraint on the intended target image, adapting framing and composition while respecting the requested edit and reference-preservation requirements. Describe any necessary changes under the changed/new-content rules. User Aspect Ratio Selection is part of the user's request. A concrete selection governs conflicting canvas wording in the editing instruction; continue to honor all compatible content constraints. Do not mechanically copy reference metadata or routing explanations into expanded_caption, and do not treat metadata as lettering inside an image.\n\nReading aspect_ratio metadata alone must not create a <Subject N> or change a preservation relationship. However, when source framing, aspect ratio, or composition is independently preserved, changed, or borrowed, identify that source-derived scope under the rules below. If realizing the selected canvas actually changes framing, visible content, placement, or layout within a Subject's defined scope, reflect that change in its relationship and describe only the changed/new consequences in Part 3. Do not call the entire source composition fully preserved while changing it; do not automatically mark every Subject partially preserved merely because the canvas ratio differs. Limit any completion or reframing to what the user's edit and selected canvas require; preserve unrelated content.\n\nRead User Aspect Ratio Selection from the final input part.\n\nSpecified ratio:\nIf a concrete ratio is supplied, copy that exact string to target_aspect_ratio. Any positive-integer W:H ratio is allowed; it is not limited to a preset list. Neither the editing instruction nor a reference image may override it.\n\nAdaptive mode:\nOnly when the selection is adaptive, determine the aspect ratio of the intended finished target image, which may differ from the main reference image. Apply these rules in order:\n\n1. Follow an explicit final target ratio or target dimensions in the editing instruction. Use the requested final geometry rather than source or earlier geometry. Exact numeric geometry takes precedence over generic orientation. When only a final orientation is specified, use 21:9 for ultra-wide, 16:9 for landscape, 1:1 for square, and 9:16 for portrait.\n\n2. Otherwise, select the main reference that supplies the edited base canvas or overall target composition and has a usable aspect_ratio. Prioritize an explicitly designated canvas/composition reference, then the edited base, then an overall composition anchor. Break ties by Picture order and skip missing, invalid, or unavailable ratios. Images used only for identity, objects, color, or style do not determine the canvas.\n\n Use the main reference's aspect_ratio as the starting ratio. Copy it exactly unless the requested edit requires a different canvas shape. Proportional resizing or changing a subject's size alone does not require changing the ratio.\n\n When expansion or cropping changes the canvas shape, determine the ratio of the resulting target. If the available information unambiguously determines it, calculate it exactly and reduce it. For relative changes:\n target_ratio = reference_ratio × width_scale / height_scale,\n where both scales are relative to the original canvas.\n\n If the requested shape change does not determine a unique ratio, choose one that fits the intended result while preserving unrelated composition. Adjust it only as needed for the requested edit, not merely for aesthetics.\n\n3. If neither an explicit target requirement nor a usable main reference determines the ratio, output 16:9.\n\nApply these rules to the whole target canvas. Grid row/column counts, individual panels, objects inside the image, and literal text are not whole-canvas ratio requests. A portrait subject or wide-angle lens alone is not a target-orientation request. Do not use a reference board's ratio as the ratio of one panel.\n\nSerialize target_aspect_ratio as two positive integers in W:H form, with no spaces; never output adaptive. Preserve concrete user selections, explicit positive-integer target ratios in the instruction, and unchanged reference ratios exactly. Convert other explicit numeric ratios or target dimensions to their exact reduced integer ratio; reduce newly calculated or chosen ratios to their lowest integer form. Do not snap to preset ratios or clamp to a supported range.\n\nLater descriptive additions must not change the chosen ratio. Pixel-size resolution, 32 alignment, and downstream size fallback are outside this task.\n\n## Interpret the User's Intent Flexibly\nRead the instruction together with the input images and any context it supplies to understand the result the user actually wants. Terse wording, colloquial phrasing, shorthand, and requests such as 'complete this' may require reasonable interpretation and substantive completion. Do not mechanically echo the request, insist that the user prewrite every detail, or treat an illustrative example or category list as the only supported case.\n\nChoose the interpretation best supported by the visible evidence, the request's purpose, and ordinary context. Supply the visual or textual details needed to make that intended result complete and coherent, including directly supporting consequences. Be flexible in how you realize the request, but do not override explicit constraints, change specified literal wording without permission, invent source facts, or expand unrelated content. Keep the boundaries between preserved source content, requested changes, newly authored text, and actual image lettering clear.\n\n## Source-Grounded Content and New Target Content\nTreat source-derived content and instruction-introduced content differently. Source facts describe what is actually visible in the inputs; target changes describe the requested result. Source fidelity does not prohibit an explicitly requested transformation. Preservation is the default only outside requested changes and their necessary consequences.\n\nA) Descriptions of source facts and properties retained in the target must be strictly faithful. Ground identity, facial features, body shape, clothing, objects, colors, materials, text, data, count, pose, layout, scene, camera, lighting, and spatial relationships in the specific input picture where they are visible. Do not invent a source attribute, accessory, logo, word, number, object, or relationship. If something is genuinely illegible or ambiguous, describe that fact neutrally rather than guessing. When an unfamiliar source element lacks a reliable name, identify it using its location and distinctive visible cues rather than guessing its identity or function.\n\nB) Content newly introduced or substantially transformed by the instruction should still be imagined richly, concretely, and coherently. Within the changed or new scope, supply plausible visual specifics that make the requested result recognizable and renderable: materials, shapes, arrangement, depth, perspective, lighting direction, shadows, reflections, contact, overlap, occlusion, atmosphere, and surrounding details that naturally belong to the requested scene. A newly requested background, environment, design, or global style may be described as a complete vivid visual scope because that scope itself is new or changed. Do not make the answer skeletal merely to satisfy a delta format.\n\nWhenever the intended target needs visual content not already determined by a preserved Subject or an explicitly referenced property, reasonably imagine and concretely describe that new content. This applies to any new scene, entity, action, design, composition, or newly visible region, not only viewpoint changes. Treat unspecified details within that new scope as room for coherent creation, not a reason to leave it vague. Keep the invention consistent with the user's intent and remaining reference constraints; do not redesign preserved Subjects or alter properties that the request requires you to match. If a source Subject gains newly visible parts, keep its identity and Subject number; describe the plausible completion in Part 3, never as an observed source fact.\n\nCreative latitude stays inside the requested or directly supporting scope. Do not redesign unrelated source content or import incidental attributes from a donor picture merely to enrich the prose. Ground factual claims in the available reference and instruction; do not invent unsupported specifics and present them as real. Explicitly requested fictional names or illustrative sample data may be authored as fictional or illustrative target content, never as recovered source facts or verified real-world information. Omit unsupported specifics, not the requested writing itself, and do not add unrelated text beyond the requested scope.\n\nComplete every textual deliverable the user asks you to write, supplement, finish, expand, rewrite, or otherwise provide. If exact wording is not supplied, compose the complete final wording yourself; missing wording is not a reason to leave the content unspecified. This applies to any requested textual deliverable, not limited to particular genres or to text inside an image. Requested wording is the deliverable itself, not forbidden restatement of the user's instruction. Do not delegate completion to the downstream model.\n\n## Output contract\nWrite expanded_caption as one continuous plain-text paragraph. Do not add structural headings, labels, bullets, markdown, preamble, reasoning, notes, or meta-commentary to the narration. Literal content may contain such words or symbols when they belong to source text or requested writing. The paragraph has three logical parts, in this order:\n1. natural definitions of the relevant source-derived <Subject N> elements;\n2. one target overview followed by a separate preservation/change relationship clause for every defined Subject;\n3. Changed and Newly Introduced Target Details Only, together with the complete wording of any textual deliverable the user asks you to provide.\n\nDetermine each requested text's destination from the instruction. If it belongs in the image, include its literal wording and describe placement as needed. If it is requested only as a textual response, provide it as text alongside the visual delta; do not turn it into a sign, caption, label, or other image element. Quoting a textual deliverable does not mean that it should be rendered in the image.\n\nComplete all Subject definitions before the target overview. Do not expose structural commands such as 'Define', 'Let', 'Identify', 'Section 1', 'Section 2', or 'Section 3' as narration. Write natural element-first definitions, for example, 'The woman in a red coat in <Picture 1> is <Subject 1>.' Refer to the intended visual result only as 'the target image'. Narrate the result, not the user's instruction or an imperative such as 'generate' or 'please create'; this does not prohibit exact supplied text or requested authored text from containing those words. If no source-derived element is retained or referenced, omit Subject definitions and relationships and begin with the target overview; do not invent a Subject to fill the structure.\n\n## Part 1: Natural Reference Element Identification\nAssign <Subject N> only to source-derived elements that remain visible or traceable in the intended target, or that provide a visible transferred or borrowed property. Every definition must name its source with an explicit <Picture N> and concise visible cues sufficient to uniquely identify the referenced object, region, or property. Add detail only to resolve ambiguity or clarify the reference/edit scope, not to restate its full appearance. Keep the full text/data transcription required below; concise identification must not omit readable content. List source pictures individually; never use ranges or vague phrases such as 'the reference images', 'the previous images', 'all inputs', or 'these pictures'. Use one number for the same convincingly identified person or element across pictures, and keep clearly different identities separate.\n\nWhen only a source property is needed, define that property rather than the entire source object; use the source object or region to locate the property. Give the source object its own Subject only if it is independently used in the target.\n\nChoose Subject granularity by relationship scope, not by a quota. An element that is independently moved, recolored, restyled, reposed, recounted, replaced while remaining traceable, or used as a donor should be independently addressable. Split a part, pose, layout, palette, lighting setup, text block, garment, accessory, held object, or background only when it needs its own preservation/change judgment. Do not mechanically split an ordinary person into face, hair, hands, every garment, and every accessory when those parts move together and have no independent role. Do not force one Subject per picture, and do not define incidental objects merely to make the inventory longer.\n\nJudge preservation over the scope actually defined. A parent Subject cannot be fully preserved if a changed child remains inside its scope, even when the child has a separate number. A genuinely unchanged narrower region or identity may be defined explicitly, but do not call the whole person, object, or scene unchanged while editing part of it. Do not transcribe the same source field again through an overlapping parent or child Subject.\n\nThe identifiable types include the following when independently relevant:\n- Entity type: a person, animal, object, creature, garment, accessory, prop, vehicle, furniture item, or other concrete entity.\n- Scene/background type: a room, landscape, street, stage, tabletop, architecture, or other environment preserved, changed, or used by the target.\n- Attribute/style type: a source-derived texture, material language, visual medium, design style, typography, border treatment, or other transferable visible quality.\n- Shared-element type: one convincingly identical visual element shown across several pictures; define it once and list every source picture.\n- Multi-form reference type: one source element used in several intended target forms, such as a physical figure, package artwork, screen render, or line drawing; define it once and distinguish the forms later.\n- Data-bearing chart/table/plot/document type: do not collapse readable data into one undifferentiated Subject. Identify the source-derived series, bars, lines, slices, columns, labeled groups, axes, titles, legends, headers, fields, and readable cells needed to preserve or edit the visible data faithfully.\n- Dense/multi-line text type: identify every clearly readable source-derived line or field in a document, slide, poster, menu, search interface, caption block, subtitle group, or UI region. Either define lines separately or define one coherent block while transcribing its lines literally and in order; do not replace readable content with a vague 'text' Subject.\n- Text/watermark type: identify an isolated watermark, brand text, subtitle, label, sign, logo text, or other short readable element, and state its clearly readable literal content once in Part 1. Later use its Subject token unless that text is intended to change.\n- Part/region type: hair, a sleeve, a wheel, a screen area, a poster corner, a body region, a panel, or another local part that needs its own relationship judgment.\n- Spatial-layout/composition type: framing, aspect ratio, grid or split-screen geometry, panel order, placement scheme, image-text hierarchy, or whitespace when the arrangement itself is preserved, changed, or borrowed.\n- Color/palette/lighting type: a source color, palette, color grade, illumination direction, warmth, contrast, shadow design, or lighting setup when color or light itself is independently referenced. Locate the specific source object or region; for example, 'The color of the cup's body in <Picture 1> is <Subject 1>.'\n- Pose/action type: a source-derived posture, gesture, gaze, body orientation, action, or interaction that is preserved, changed, or transferred.\n- Logo/symbol/icon type: a graphical emblem, symbol, UI icon, or pictographic mark; handle any readable words under the text rules.\n- Embedded/thumbnail type: a complete or cropped source picture embedded in the target as a panel or thumbnail.\n- Partial-crop type: a target crop or zoom whose visible content comes from a specific region of a source picture.\n\nThese fifteen types form the complete supported taxonomy, not a per-example checklist or quota. Use every applicable type supported by the input images and the instruction when it needs an independent relationship judgment, but never create an inapplicable Subject merely to instantiate a type. The more specific part/layout/light/pose/icon types exist to prevent an independently changed unit from being buried inside an oversized Subject, not to force mechanical over-splitting.\n\nBrand-new target-only content has no <Subject N>. A source element explicitly removed or wholly replaced and not otherwise reused also has no Subject number; if necessary, identify it once with its <Picture N> in the target overview, then describe the positive completed region in Part 3. Editing or translating a source text field is a traceable change to that field, not removal of the entire element: retain its Subject when needed to locate the edit, record its original wording in Part 1, and give its full resulting wording in Part 3. Every defined Subject must receive a relationship later and must not be defined merely to say that it is absent, removed, unused, or forgotten.\n\nReadable source text and data require special care. For every defined source-derived chart, table, document, UI, text block, or isolated text element, transcribe exactly once in Part 1 every clearly readable source-state line, character, label, value, axis title, axis tick, legend entry, table header, field, and readable cell, including a source literal that the instruction intends to change. Preserve the original language, script, grouping, and reading order. Do not omit a clearly readable value merely because it is unchanged or peripheral to the main edit. Never guess hidden, cropped-away, or illegible source text.\n\nFor grouped or clustered bars, radar charts, multi-panel plots, and many-point plots, enumerate every clearly readable value rather than summarizing the display. Check counts only when the chart's visible semantics require them: grouped bars against the legend, radar values against the axes, and mutually exclusive percentage parts of a complete whole against approximately 100 percent. Do not impose those assumptions on unrelated chart types or incomplete, overlapping, or multi-select categories. If the source itself contains a genuine inconsistency, missing series, or illegible value, record only the visible evidence or the uncertainty in Part 1; never silently normalize or repair the source record. An explicitly requested correction belongs in the target changes and must use supplied or derivable values, not guessed real data. Exhaustive transcription does not require one Subject per value or line: a coherent series, row, field group, or text block may remain one Subject when its literals are fully enumerated and it does not need separate relationship judgments. These checks apply only where the relevant marks and values are actually readable, and they do not require turning an ordinary non-data image into a chart inventory.\n\n## Part 2: Target Overview and Reference Relationships\nAfter the definitions, write one natural sentence beginning with 'The target image'. Foreground the primary intended changed or new state and describe the overall result using the defined Subject tokens where appropriate. Target-only new content remains unnumbered. The overview may summarize several requested changes, but it should remain an overview rather than a second detailed caption.\n\nThen give every defined Subject exactly one whole-subject relationship in its own independent clause. Give each Subject its own relationship clause; do not combine several Subjects into one relationship statement, even when they share the same degree. Keep the prose natural without imposing a fixed sentence length, a fixed Subject count, or a target frequency for any relationship type. Use these four relationships:\n\n- Fully preserved: use only when the whole Subject is unchanged in identity, appearance, count, pose or state, position, topology, and intended target role. Write only the short standalone clause '<Subject N> is fully preserved.' Give each fully-preserved Subject its own sentence; never merge multiple Subjects into one fully-preserved clause. Do not append a checklist, explanation, source citation, or visual description to a fully-preserved clause.\n- Partially preserved: use when the Subject remains traceable but at least one relevant property changes. State the retained categories needed to anchor identity and the changed categories, including a concrete before-to-after relationship when the source and instruction determine it. Do not exhaustively re-render retained appearance.\n- Attribute transfer: use when a specific source-derived property is directly reused or matched on a target carrier. Name the source Subject, the transferred property, and the target carrier. Transfer only the assigned property, not unrelated donor attributes.\n- Weak reference: use when the source guides a general style, palette, atmosphere, composition principle, or similar quality without requiring a specific property to be matched exactly. Name the borrowed quality and make clear what the new design may vary. Choose by the requested use of the reference, not merely by whether the property is called a color, style, or palette.\n\nAttribute-transfer example: the user wants the coat in <Picture 2> to match the cup's body color in <Picture 1>. Part 1 defines that color as <Subject 1> and the coat as <Subject 2>. The relationship clauses can be '<Subject 1> supplies the exact color transferred to <Subject 2>. <Subject 2> is partially preserved; its color changes while its cut and material remain.' The cup's shape and material are not transferred.\n\nWeak-reference example: the user wants a new abstract composition inspired by a landscape's palette, without copying it exactly. Part 1 defines the landscape's palette as <Subject 1>. The relationship clause can be '<Subject 1> serves as a weak palette reference for the new abstract composition; related colors are redistributed among geometric blocks, with freely adapted proportions and placement.' The new composition has no Subject number. These examples illustrate the distinction, not a restriction of either relationship to color tasks.\n\nFor a Subject used in several target forms or instances, judge the intended uses together. A mix of unchanged and transformed or additional instances is partially preserved overall; distinguish those uses within that Subject's own relationship statement. Do not call the entire Subject fully preserved just because one instance is unchanged. If only a property is borrowed, use the appropriate transfer or weak-reference relationship. If a donor object remains unchanged while its color or another property is transferred elsewhere, define that property separately so the object's preservation and the property's transfer have distinct scopes.\n\nThe relationship must agree with the described target. If a requested change or its necessary integration changes a Subject's position, orientation, count, topology, role, pose, expression, clothing, color, material, lighting, state, or other property within its defined scope, the whole Subject is not fully preserved. If later changed-detail prose genuinely needs to describe a change to a Subject, classify that Subject as partially preserved instead. A source carrier may remain fully preserved while one of its properties is transferred outward; do not mark the carrier partial unless the carrier itself changes. Conversely, new shadows, reflections, occlusion, or relighting on the carrier count as changes where they affect its defined appearance; do not hide them behind a fully-preserved judgment.\n\nFor partially preserved chart, table, document, UI, and text-block Subjects, state which categories are retained and which change. For a fully preserved chart or text Subject, use only the standard fully-preserved sentence, not an extra checklist. Keep relationship statements compact: use Part 1 as the source-side record and leave complete changed or newly authored wording and values for Part 3. Do not duplicate old literals in a before-to-after phrase or new literals in both the relationship and detail blocks.\n\nTreat unmentioned source-derived content as stable by default, but do not let that conservative baseline suppress the requested imagination. The changed/new target scope may still be rich. Direct semantic, spatial, physical, compositional, environmental, camera, and lighting consequences of the requested change are part of the changed scope and may be described.\n\n## Part 3: Changed and Newly Introduced Target Details Only\nThis final logical block is a richly rendered visual delta, not a standalone T2I re-caption of the whole target image. Also supply the requested textual deliverables. For visual description, cover only:\n- the concrete final appearance of source-derived attributes that actually change;\n- target-only people, objects, text regions, environments, panels, charts, layouts, effects, or other content introduced by the instruction;\n- the newly exposed or positively completed surface or space left by a removal, or the positive new occupant of a replacement;\n- plausible supporting details inside the changed/new scope that make it visually recognizable and coherent;\n- direct integration consequences involving scale, orientation, contact, overlap, occlusion, perspective, depth, shadow, reflection, lighting, material response, anatomy, support, camera framing, or composition.\n\nThis changed/new-only rule does not impose terse writing. A newly introduced restaurant, fantasy landscape, product design, comic panel, interface, lighting scheme, or global style may be described vividly and comprehensively because that entire requested scope is new or changed. Likewise, a transformed garment, pose, object, chart, or layout may receive enough concrete detail to be rendered reliably. Be imaginative where the instruction creates room for imagination; simply keep that imagination inside the changed/new scope instead of using it to re-caption preserved source content.\n\nStrict non-repetition rules:\n- After a Subject is classified as fully preserved, never visually describe that Subject again. A later sentence may use only the bare <Subject N> token as the minimum spatial or interaction anchor needed to locate changed or new content, for example, 'A new luminous pearl rests beside <Subject 2>.' Do not attach an adjective, color, material, shape, part, pose, position, role, or other appearance detail to that preserved Subject.\n- For a partially preserved Subject, Part 3 describes only its changed attributes and their concrete intended final appearance. Do not repeat the detailed appearance of retained identity, clothing, body, object, scene, or layout attributes already grounded in Part 1 or named as retained in Part 2.\n- For attribute transfer or weak reference, describe only how the transferred or borrowed quality is realized on the intended target carrier. Do not import or re-render unrelated donor properties.\n- Do not repeat fully preserved source text or data as visual description; use the Part 1 record and bare <Subject N> references. If the user separately requests a complete transcription or quotation as a textual deliverable, write that deliverable in full even when it repeats source wording; this is required output content, not redundant visual description.\n- Write every changed or target-only text/data field and requested textual deliverable in full. Use exact supplied or deterministically derived results; preserve every supplied character within each line, including punctuation, capitalization, order, and language, unless the user explicitly requests translation or another textual transformation, in which case write its actual result. If a line is edited, write the complete resulting line, not merely the changed word or numeric difference; do not recopy separate unaffected lines as visual description. Never substitute a <Subject N> token, a description or writing instruction, placeholders, omitted middles, 'etc.', 'and so on', or ellipses for requested content; preserve an ellipsis that belongs to supplied literal text.\n- Do not describe an unchanged background, person, object, pose, camera, light, style, or image-quality property merely to make the paragraph resemble a complete T2I caption.\n- If no changed or newly introduced detail remains beyond the overview and relationship statements and no requested textual deliverable remains to be supplied, stop. Never pad the output with preserved appearance.\n\nFor example, for a reference showing a lakeside path and hills, a request to write a scenic introduction requires actual copy such as '沿着湖畔步道缓缓前行，欣赏湖水与远山相映的景色，在悠闲漫步中感受旅途的宁静。', not 'a paragraph introducing the attraction'. This illustrates completion and is not a restriction to travel copy or permission to reuse these words in unrelated tasks.\n\nCorrect pattern: 'The black horse in <Picture 1> is <Subject 1>. The target image shows <Subject 1> beside a newly added luminous pearl. <Subject 1> is fully preserved. The pearl emits a cool glow onto the ground beside <Subject 1>.' The final sentence describes the new pearl and its integration while using the bare Subject only as an anchor.\n\nIncorrect pattern: '<Subject 1> is fully preserved. <Subject 1> is a muscular black horse with segmented dark armor and a long mane.' The second sentence is a T2I-style re-caption of preserved content and must be omitted.\n\nCorrect partial pattern: '<Subject 1> is partially preserved; its identity and body armor remain, while its bare head changes to an ornate helmet. The new helmet has a pointed silver crest, a red plume, and shadows that follow the existing head orientation.' Do not subsequently re-describe the retained face, body, or armor.\n\nFor exact reference-color or palette matching, use the defined color or palette <Subject N> to specify where it is applied on the target. The visible reference, not an approximate color name, determines the color; do not replace it with inferred color names or guessed RGB/HEX values. Honor explicit user-supplied colors or values, and do not transfer unrelated donor material or lighting unless requested.\n\nWhen describing a grid, multi-panel image, storyboard, table-like arrangement, or other row-and-column layout, state the row count and column count explicitly, either in words such as '2 rows and 3 columns' or with labeled multiplication notation such as '2 x 3 (rows x columns)' or '3 x 2 (columns x rows)'. Multiplication notation is allowed; specify which number denotes rows and which denotes columns. Do not use an unlabeled '2 x 3' or '2×3' alone. When relevant, describe the cell contents in top-to-bottom row order and left-to-right column order, and name which row and column contains each changed or newly introduced element.\n\nFor uneven rows, merged cells, or a freeform composition, describe the actual arrangement directly, such as 'the top row has three panels and the bottom row has two', with positions and spans as needed; do not impose a rectangular grid that is not intended. If shorthand in the instruction is ambiguous, use the reference and context to interpret the intended arrangement and express the chosen row/column meaning explicitly. Record relevant unchanged source layout in the Subject definitions without repeating it after a fully-preserved judgment; Part 3 describes changed or new layout only. This wording rule does not authorize altering supplied literal text containing dimension notation.\n\n## Final language and self-check\nWrite all descriptive and relationship prose in fluent English, regardless of the language of the user instruction. For newly authored requested copy, use the user's requested language, or otherwise the language of the user's request. Keep literal text and completed requested textual deliverables in single quotes to distinguish them from the surrounding English visual description and relationship prose, preserving apostrophes and other punctuation inside the literal text. Represent multi-line literal content inline as separate quoted lines in reading order, describing the intended line breaks or layout in prose rather than inserting actual newline characters into the response.\n\nUse objective, concrete visual language. Richness should come from precise changed/new appearance and coherent integration, not from repetition. Describe visible expressions rather than asserting unobservable thoughts or feelings. Preserve expressions outside the changed scope; expressions for newly introduced or transformed people may be reasonably imagined within the request. Write reference markers <Picture N> and <Subject N> directly without backticks; do not rewrite a literal text string merely because it contains similar words or symbols.\n\nBefore answering, verify:\n- every source visual fact is visible in a named <Picture N>, and every intended change is supported by the editing instruction or User Aspect Ratio Selection and their necessary consequences; reference aspect ratios come only from the supplied metadata;\n- the output still imagines instruction-introduced content richly enough to be recognizable and renderable;\n- every Subject definition is concise but unambiguous about its source and reference scope, without unnecessary appearance detail or missing required text/data;\n- every Subject is naturally defined before use and has its own independent relationship clause, never merged with another Subject;\n- every fully preserved Subject has its own exact '<Subject N> is fully preserved.' sentence and receives no later visual re-description;\n- no fully preserved parent contains an edited child, and mixed target forms or integration effects are reflected in the relationship;\n- every partially preserved Subject repeats no detailed retained appearance in Part 3;\n- donor attributes do not leak beyond the property requested;\n- target-only and removed-only content has no Subject number;\n- every applicable reference-element type is supported without treating the taxonomy as a per-example quota;\n- source text/data is recorded once in Part 1 without guessing, repairing, or translating the source record;\n- all requested target text and textual deliverables are complete and literal, with source repetition only for an explicitly requested transcription or quotation;\n- each textual deliverable has the requested image or text-only destination;\n- Part 3's visual description contains changed/new content and direct integration only, with no whole-frame T2I padding;\n- all non-literal descriptive and relationship prose is English, while quoted source, user-specified, or newly authored requested text uses its required language and script;\n- any grid layout described makes the row count and column count unambiguous, either in words or with explicitly labeled multiplication notation, with positions or uneven-row structure clarified where needed;\n- expanded_caption is one fluent, complete paragraph with no structural meta-language in the narration; multi-line literals retain their wording, reading order, and intended line boundaries through inline quoted lines;\n- target_aspect_ratio copies any concrete user selection exactly; in adaptive mode, it preserves the main reference ratio when no canvas-shape change is required and reflects any requested expansion or cropping; expanded_caption and its preservation relationships agree with the final canvas;\n- the final answer is a valid JSON object containing only the two string fields target_aspect_ratio and expanded_caption.\n\nOutput only the required two-field JSON object.\n\nReference Image Count:\n{d} input image(s), labeled <Picture 1> through <Picture {d}>.\n\n{a:.{b}}\nUser Editing Instruction (JSON string):\n\"{c}\"\n\nUser Aspect Ratio Selection:\nadaptive\n\nApply the instructions above to the supplied inputs and reference images. Output override: the canvas is already fixed. Write only the expanded_caption text itself: one fluent English paragraph. No JSON, no keys, no braces, no Markdown, nothing before or after the paragraph.<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n"
4217 },
4218 "color": "#14424d",
4219 "bgcolor": "#161f21"
4220 },
4221 {
4222 "id": 410,
4223 "type": "AUSBOSS_NODES_Callout",
4224 "pos": [
4225 1100,
4226 80
4227 ],
4228 "size": [
4229 460,
4230 150
4231 ],
4232 "flags": {
4233 "pinned": true
4234 },
4235 "order": 3,
4236 "mode": 0,
4237 "inputs": [
4238 {
4239 "localized_name": "text",
4240 "name": "text",
4241 "type": "STRING",
4242 "widget": {
4243 "name": "text"
4244 },
4245 "link": null
4246 }
4247 ],
4248 "outputs": [],
4249 "title": "Callout",
4250 "properties": {
4251 "Node name for S&R": "AUSBOSS_NODES_Callout",
4252 "cnr_id": "ausboss-nodes"
4253 },
4254 "widgets_values": [
4255 "The writer is Qwen3.5 9B, the one Comfy recommends for this model. Thinking is off.\n###### Same words and pictures give the same prompt."
4256 ],
4257 "widgets_values_named": {
4258 "text": "The writer is Qwen3.5 9B, the one Comfy recommends for this model. Thinking is off.\n###### Same words and pictures give the same prompt."
4259 },
4260 "color": "#07090a",
4261 "bgcolor": "#07090a"
4262 },
4263 {
4264 "id": 411,
4265 "type": "TextGenerate",
4266 "pos": [
4267 1100,
4268 300
4269 ],
4270 "size": [
4271 460,
4272 300
4273 ],
4274 "flags": [],
4275 "order": 11,
4276 "mode": 0,
4277 "inputs": [
4278 {
4279 "localized_name": "clip",
4280 "name": "clip",
4281 "type": "CLIP",
4282 "link": 1048
4283 },
4284 {
4285 "localized_name": "image",
4286 "name": "image",
4287 "shape": 7,
4288 "type": "IMAGE",
4289 "link": null
4290 },
4291 {
4292 "localized_name": "video",
4293 "name": "video",
4294 "shape": 7,
4295 "type": "IMAGE",
4296 "link": null
4297 },
4298 {
4299 "localized_name": "audio",
4300 "name": "audio",
4301 "shape": 7,
4302 "type": "AUDIO",
4303 "link": null
4304 },
4305 {
4306 "localized_name": "system_prompt",
4307 "name": "system_prompt",
4308 "shape": 7,
4309 "type": "STRING",
4310 "link": null
4311 },
4312 {
4313 "localized_name": "prompt",
4314 "name": "prompt",
4315 "type": "STRING",
4316 "widget": {
4317 "name": "prompt"
4318 },
4319 "link": 1050
4320 },
4321 {
4322 "localized_name": "max_length",
4323 "name": "max_length",
4324 "type": "INT",
4325 "widget": {
4326 "name": "max_length"
4327 },
4328 "link": null
4329 },
4330 {
4331 "localized_name": "Sampling Mode",
4332 "name": "sampling_mode",
4333 "type": "COMFY_DYNAMICCOMBO_V3",
4334 "widget": {
4335 "name": "sampling_mode"
4336 },
4337 "link": null
4338 },
4339 {
4340 "localized_name": "thinking",
4341 "name": "thinking",
4342 "shape": 7,
4343 "type": "BOOLEAN",
4344 "widget": {
4345 "name": "thinking"
4346 },
4347 "link": null
4348 },
4349 {
4350 "localized_name": "use_default_template",
4351 "name": "use_default_template",
4352 "shape": 7,
4353 "type": "BOOLEAN",
4354 "widget": {
4355 "name": "use_default_template"
4356 },
4357 "link": null
4358 },
4359 {
4360 "localized_name": "mtp",
4361 "name": "mtp",
4362 "shape": 7,
4363 "type": "COMBO",
4364 "widget": {
4365 "name": "mtp"
4366 },
4367 "link": null
4368 }
4369 ],
4370 "outputs": [
4371 {
4372 "localized_name": "generated_text",
4373 "name": "generated_text",
4374 "type": "STRING",
4375 "links": [
4376 1053
4377 ]
4378 },
4379 {
4380 "localized_name": "thinking",
4381 "name": "thinking",
4382 "type": "STRING",
4383 "links": []
4384 }
4385 ],
4386 "title": "Generate Text - The writer, for a new picture",
4387 "properties": {
4388 "Node name for S&R": "TextGenerate",
4389 "cnr_id": "comfy-core",
4390 "ver": "0.44.0"
4391 },
4392 "widgets_values": [
4393 "",
4394 2048,
4395 "off",
4396 false,
4397 true,
4398 "auto"
4399 ],
4400 "widgets_values_named": {
4401 "prompt": "",
4402 "max_length": 2048,
4403 "sampling_mode": "off",
4404 "thinking": false,
4405 "use_default_template": true,
4406 "mtp": "auto"
4407 },
4408 "color": "#14424d",
4409 "bgcolor": "#161f21"
4410 },
4411 {
4412 "id": 412,
4413 "type": "TextGenerate",
4414 "pos": [
4415 1100,
4416 670
4417 ],
4418 "size": [
4419 460,
4420 300
4421 ],
4422 "flags": [],
4423 "order": 12,
4424 "mode": 0,
4425 "inputs": [
4426 {
4427 "localized_name": "clip",
4428 "name": "clip",
4429 "type": "CLIP",
4430 "link": 1049
4431 },
4432 {
4433 "localized_name": "image",
4434 "name": "image",
4435 "shape": 7,
4436 "type": "IMAGE",
4437 "link": 1052
4438 },
4439 {
4440 "localized_name": "video",
4441 "name": "video",
4442 "shape": 7,
4443 "type": "IMAGE",
4444 "link": null
4445 },
4446 {
4447 "localized_name": "audio",
4448 "name": "audio",
4449 "shape": 7,
4450 "type": "AUDIO",
4451 "link": null
4452 },
4453 {
4454 "localized_name": "system_prompt",
4455 "name": "system_prompt",
4456 "shape": 7,
4457 "type": "STRING",
4458 "link": null
4459 },
4460 {
4461 "localized_name": "prompt",
4462 "name": "prompt",
4463 "type": "STRING",
4464 "widget": {
4465 "name": "prompt"
4466 },
4467 "link": 1051
4468 },
4469 {
4470 "localized_name": "max_length",
4471 "name": "max_length",
4472 "type": "INT",
4473 "widget": {
4474 "name": "max_length"
4475 },
4476 "link": null
4477 },
4478 {
4479 "localized_name": "Sampling Mode",
4480 "name": "sampling_mode",
4481 "type": "COMFY_DYNAMICCOMBO_V3",
4482 "widget": {
4483 "name": "sampling_mode"
4484 },
4485 "link": null
4486 },
4487 {
4488 "localized_name": "thinking",
4489 "name": "thinking",
4490 "shape": 7,
4491 "type": "BOOLEAN",
4492 "widget": {
4493 "name": "thinking"
4494 },
4495 "link": null
4496 },
4497 {
4498 "localized_name": "use_default_template",
4499 "name": "use_default_template",
4500 "shape": 7,
4501 "type": "BOOLEAN",
4502 "widget": {
4503 "name": "use_default_template"
4504 },
4505 "link": null
4506 },
4507 {
4508 "localized_name": "mtp",
4509 "name": "mtp",
4510 "shape": 7,
4511 "type": "COMBO",
4512 "widget": {
4513 "name": "mtp"
4514 },
4515 "link": null
4516 }
4517 ],
4518 "outputs": [
4519 {
4520 "localized_name": "generated_text",
4521 "name": "generated_text",
4522 "type": "STRING",
4523 "links": [
4524 1054
4525 ]
4526 },
4527 {
4528 "localized_name": "thinking",
4529 "name": "thinking",
4530 "type": "STRING",
4531 "links": []
4532 }
4533 ],
4534 "title": "Generate Text - The writer, for an edit",
4535 "properties": {
4536 "Node name for S&R": "TextGenerate",
4537 "cnr_id": "comfy-core",
4538 "ver": "0.44.0"
4539 },
4540 "widgets_values": [
4541 "",
4542 2048,
4543 "off",
4544 false,
4545 true,
4546 "auto"
4547 ],
4548 "widgets_values_named": {
4549 "prompt": "",
4550 "max_length": 2048,
4551 "sampling_mode": "off",
4552 "thinking": false,
4553 "use_default_template": true,
4554 "mtp": "auto"
4555 },
4556 "color": "#14424d",
4557 "bgcolor": "#161f21"
4558 },
4559 {
4560 "id": 413,
4561 "type": "ComfySwitchNode",
4562 "pos": [
4563 1100,
4564 1040
4565 ],
4566 "size": [
4567 460,
4568 78
4569 ],
4570 "flags": [],
4571 "order": 13,
4572 "mode": 0,
4573 "inputs": [
4574 {
4575 "localized_name": "on_false",
4576 "name": "on_false",
4577 "shape": 7,
4578 "type": "STRING",
4579 "link": 1053
4580 },
4581 {
4582 "localized_name": "on_true",
4583 "name": "on_true",
4584 "shape": 7,
4585 "type": "STRING",
4586 "link": 1054
4587 },
4588 {
4589 "localized_name": "switch",
4590 "name": "switch",
4591 "type": "BOOLEAN",
4592 "widget": {
4593 "name": "switch"
4594 },
4595 "link": 1055
4596 }
4597 ],
4598 "outputs": [
4599 {
4600 "localized_name": "output",
4601 "name": "output",
4602 "type": "STRING",
4603 "links": [
4604 1143
4605 ]
4606 }
4607 ],
4608 "title": "If/Else Switch - New picture or edit",
4609 "properties": {
4610 "Node name for S&R": "ComfySwitchNode",
4611 "cnr_id": "comfy-core",
4612 "ver": "0.44.0"
4613 },
4614 "widgets_values": [
4615 true
4616 ],
4617 "widgets_values_named": {
4618 "switch": true
4619 },
4620 "color": "#14424d",
4621 "bgcolor": "#161f21"
4622 },
4623 {
4624 "id": 420,
4625 "type": "AUSBOSS_NODES_Callout",
4626 "pos": [
4627 1620,
4628 80
4629 ],
4630 "size": [
4631 460,
4632 150
4633 ],
4634 "flags": {
4635 "pinned": true
4636 },
4637 "order": 4,
4638 "mode": 0,
4639 "inputs": [
4640 {
4641 "localized_name": "text",
4642 "name": "text",
4643 "type": "STRING",
4644 "widget": {
4645 "name": "text"
4646 },
4647 "link": null
4648 }
4649 ],
4650 "outputs": [],
4651 "title": "Callout",
4652 "properties": {
4653 "Node name for S&R": "AUSBOSS_NODES_Callout",
4654 "cnr_id": "ausboss-nodes"
4655 },
4656 "widgets_values": [
4657 "Your extra pictures go on to the box that draws.\n###### An empty socket passes nothing on."
4658 ],
4659 "widgets_values_named": {
4660 "text": "Your extra pictures go on to the box that draws.\n###### An empty socket passes nothing on."
4661 },
4662 "color": "#07090a",
4663 "bgcolor": "#07090a"
4664 },
4665 {
4666 "id": 421,
4667 "type": "ComfySwitchNode",
4668 "pos": [
4669 1620,
4670 300
4671 ],
4672 "size": [
4673 460,
4674 78
4675 ],
4676 "flags": [],
4677 "order": 15,
4678 "mode": 0,
4679 "inputs": [
4680 {
4681 "localized_name": "on_false",
4682 "name": "on_false",
4683 "shape": 7,
4684 "type": "IMAGE",
4685 "link": null
4686 },
4687 {
4688 "localized_name": "on_true",
4689 "name": "on_true",
4690 "shape": 7,
4691 "type": "*",
4692 "link": 1057
4693 },
4694 {
4695 "localized_name": "switch",
4696 "name": "switch",
4697 "type": "BOOLEAN",
4698 "widget": {
4699 "name": "switch"
4700 },
4701 "link": null
4702 }
4703 ],
4704 "outputs": [
4705 {
4706 "localized_name": "output",
4707 "name": "output",
4708 "type": "IMAGE",
4709 "links": [
4710 1058
4711 ]
4712 }
4713 ],
4714 "title": "If/Else Switch - Picture 2 on",
4715 "properties": {
4716 "Node name for S&R": "ComfySwitchNode",
4717 "cnr_id": "comfy-core",
4718 "ver": "0.44.0"
4719 },
4720 "widgets_values": [
4721 true
4722 ],
4723 "widgets_values_named": {
4724 "switch": true
4725 },
4726 "color": "#14424d",
4727 "bgcolor": "#161f21"
4728 },
4729 {
4730 "id": 422,
4731 "type": "ComfySwitchNode",
4732 "pos": [
4733 1620,
4734 420
4735 ],
4736 "size": [
4737 460,
4738 78
4739 ],
4740 "flags": [],
4741 "order": 16,
4742 "mode": 0,
4743 "inputs": [
4744 {
4745 "localized_name": "on_false",
4746 "name": "on_false",
4747 "shape": 7,
4748 "type": "IMAGE",
4749 "link": null
4750 },
4751 {
4752 "localized_name": "on_true",
4753 "name": "on_true",
4754 "shape": 7,
4755 "type": "*",
4756 "link": 1059
4757 },
4758 {
4759 "localized_name": "switch",
4760 "name": "switch",
4761 "type": "BOOLEAN",
4762 "widget": {
4763 "name": "switch"
4764 },
4765 "link": null
4766 }
4767 ],
4768 "outputs": [
4769 {
4770 "localized_name": "output",
4771 "name": "output",
4772 "type": "IMAGE",
4773 "links": [
4774 1060
4775 ]
4776 }
4777 ],
4778 "title": "If/Else Switch - Picture 3 on",
4779 "properties": {
4780 "Node name for S&R": "ComfySwitchNode",
4781 "cnr_id": "comfy-core",
4782 "ver": "0.44.0"
4783 },
4784 "widgets_values": [
4785 true
4786 ],
4787 "widgets_values_named": {
4788 "switch": true
4789 },
4790 "color": "#14424d",
4791 "bgcolor": "#161f21"
4792 },
4793 {
4794 "id": 423,
4795 "type": "ComfySwitchNode",
4796 "pos": [
4797 1620,
4798 540
4799 ],
4800 "size": [
4801 460,
4802 78
4803 ],
4804 "flags": [],
4805 "order": 17,
4806 "mode": 0,
4807 "inputs": [
4808 {
4809 "localized_name": "on_false",
4810 "name": "on_false",
4811 "shape": 7,
4812 "type": "IMAGE",
4813 "link": null
4814 },
4815 {
4816 "localized_name": "on_true",
4817 "name": "on_true",
4818 "shape": 7,
4819 "type": "*",
4820 "link": 1061
4821 },
4822 {
4823 "localized_name": "switch",
4824 "name": "switch",
4825 "type": "BOOLEAN",
4826 "widget": {
4827 "name": "switch"
4828 },
4829 "link": null
4830 }
4831 ],
4832 "outputs": [
4833 {
4834 "localized_name": "output",
4835 "name": "output",
4836 "type": "IMAGE",
4837 "links": [
4838 1062
4839 ]
4840 }
4841 ],
4842 "title": "If/Else Switch - Picture 4 on",
4843 "properties": {
4844 "Node name for S&R": "ComfySwitchNode",
4845 "cnr_id": "comfy-core",
4846 "ver": "0.44.0"
4847 },
4848 "widgets_values": [
4849 true
4850 ],
4851 "widgets_values_named": {
4852 "switch": true
4853 },
4854 "color": "#14424d",
4855 "bgcolor": "#161f21"
4856 },
4857 {
4858 "id": 424,
4859 "type": "ComfySwitchNode",
4860 "pos": [
4861 1620,
4862 660
4863 ],
4864 "size": [
4865 460,
4866 78
4867 ],
4868 "flags": [],
4869 "order": 18,
4870 "mode": 0,
4871 "inputs": [
4872 {
4873 "localized_name": "on_false",
4874 "name": "on_false",
4875 "shape": 7,
4876 "type": "IMAGE",
4877 "link": null
4878 },
4879 {
4880 "localized_name": "on_true",
4881 "name": "on_true",
4882 "shape": 7,
4883 "type": "*",
4884 "link": 1063
4885 },
4886 {
4887 "localized_name": "switch",
4888 "name": "switch",
4889 "type": "BOOLEAN",
4890 "widget": {
4891 "name": "switch"
4892 },
4893 "link": null
4894 }
4895 ],
4896 "outputs": [
4897 {
4898 "localized_name": "output",
4899 "name": "output",
4900 "type": "IMAGE",
4901 "links": [
4902 1064
4903 ]
4904 }
4905 ],
4906 "title": "If/Else Switch - Picture 5 on",
4907 "properties": {
4908 "Node name for S&R": "ComfySwitchNode",
4909 "cnr_id": "comfy-core",
4910 "ver": "0.44.0"
4911 },
4912 "widgets_values": [
4913 true
4914 ],
4915 "widgets_values_named": {
4916 "switch": true
4917 },
4918 "color": "#14424d",
4919 "bgcolor": "#161f21"
4920 },
4921 {
4922 "id": 425,
4923 "type": "ComfySwitchNode",
4924 "pos": [
4925 1620,
4926 780
4927 ],
4928 "size": [
4929 460,
4930 78
4931 ],
4932 "flags": [],
4933 "order": 19,
4934 "mode": 0,
4935 "inputs": [
4936 {
4937 "localized_name": "on_false",
4938 "name": "on_false",
4939 "shape": 7,
4940 "type": "IMAGE",
4941 "link": null
4942 },
4943 {
4944 "localized_name": "on_true",
4945 "name": "on_true",
4946 "shape": 7,
4947 "type": "*",
4948 "link": 1065
4949 },
4950 {
4951 "localized_name": "switch",
4952 "name": "switch",
4953 "type": "BOOLEAN",
4954 "widget": {
4955 "name": "switch"
4956 },
4957 "link": null
4958 }
4959 ],
4960 "outputs": [
4961 {
4962 "localized_name": "output",
4963 "name": "output",
4964 "type": "IMAGE",
4965 "links": [
4966 1066
4967 ]
4968 }
4969 ],
4970 "title": "If/Else Switch - Picture 6 on",
4971 "properties": {
4972 "Node name for S&R": "ComfySwitchNode",
4973 "cnr_id": "comfy-core",
4974 "ver": "0.44.0"
4975 },
4976 "widgets_values": [
4977 true
4978 ],
4979 "widgets_values_named": {
4980 "switch": true
4981 },
4982 "color": "#14424d",
4983 "bgcolor": "#161f21"
4984 },
4985 {
4986 "id": 426,
4987 "type": "ComfySwitchNode",
4988 "pos": [
4989 1620,
4990 900
4991 ],
4992 "size": [
4993 460,
4994 78
4995 ],
4996 "flags": [],
4997 "order": 20,
4998 "mode": 0,
4999 "inputs": [
5000 {
5001 "localized_name": "on_false",
5002 "name": "on_false",
5003 "shape": 7,
5004 "type": "IMAGE",
5005 "link": null
5006 },
5007 {
5008 "localized_name": "on_true",
5009 "name": "on_true",
5010 "shape": 7,
5011 "type": "*",
5012 "link": 1067
5013 },
5014 {
5015 "localized_name": "switch",
5016 "name": "switch",
5017 "type": "BOOLEAN",
5018 "widget": {
5019 "name": "switch"
5020 },
5021 "link": null
5022 }
5023 ],
5024 "outputs": [
5025 {
5026 "localized_name": "output",
5027 "name": "output",
5028 "type": "IMAGE",
5029 "links": [
5030 1068
5031 ]
5032 }
5033 ],
5034 "title": "If/Else Switch - Picture 7 on",
5035 "properties": {
5036 "Node name for S&R": "ComfySwitchNode",
5037 "cnr_id": "comfy-core",
5038 "ver": "0.44.0"
5039 },
5040 "widgets_values": [
5041 true
5042 ],
5043 "widgets_values_named": {
5044 "switch": true
5045 },
5046 "color": "#14424d",
5047 "bgcolor": "#161f21"
5048 },
5049 {
5050 "id": 427,
5051 "type": "ComfySwitchNode",
5052 "pos": [
5053 1620,
5054 1020
5055 ],
5056 "size": [
5057 460,
5058 78
5059 ],
5060 "flags": [],
5061 "order": 21,
5062 "mode": 0,
5063 "inputs": [
5064 {
5065 "localized_name": "on_false",
5066 "name": "on_false",
5067 "shape": 7,
5068 "type": "IMAGE",
5069 "link": null
5070 },
5071 {
5072 "localized_name": "on_true",
5073 "name": "on_true",
5074 "shape": 7,
5075 "type": "*",
5076 "link": 1069
5077 },
5078 {
5079 "localized_name": "switch",
5080 "name": "switch",
5081 "type": "BOOLEAN",
5082 "widget": {
5083 "name": "switch"
5084 },
5085 "link": null
5086 }
5087 ],
5088 "outputs": [
5089 {
5090 "localized_name": "output",
5091 "name": "output",
5092 "type": "IMAGE",
5093 "links": [
5094 1070
5095 ]
5096 }
5097 ],
5098 "title": "If/Else Switch - Picture 8 on",
5099 "properties": {
5100 "Node name for S&R": "ComfySwitchNode",
5101 "cnr_id": "comfy-core",
5102 "ver": "0.44.0"
5103 },
5104 "widgets_values": [
5105 true
5106 ],
5107 "widgets_values_named": {
5108 "switch": true
5109 },
5110 "color": "#14424d",
5111 "bgcolor": "#161f21"
5112 },
5113 {
5114 "id": 414,
5115 "type": "ComfySwitchNode",
5116 "pos": [
5117 560,
5118 328
5119 ],
5120 "size": [
5121 460,
5122 78
5123 ],
5124 "flags": [],
5125 "order": 14,
5126 "mode": 0,
5127 "inputs": [
5128 {
5129 "localized_name": "on_false",
5130 "name": "on_false",
5131 "shape": 7,
5132 "type": "IMAGE",
5133 "link": null
5134 },
5135 {
5136 "localized_name": "on_true",
5137 "name": "on_true",
5138 "shape": 7,
5139 "type": "*",
5140 "link": 1034
5141 },
5142 {
5143 "localized_name": "switch",
5144 "name": "switch",
5145 "type": "BOOLEAN",
5146 "widget": {
5147 "name": "switch"
5148 },
5149 "link": 1071
5150 }
5151 ],
5152 "outputs": [
5153 {
5154 "localized_name": "output",
5155 "name": "output",
5156 "type": "IMAGE",
5157 "links": [
5158 1072
5159 ]
5160 }
5161 ],
5162 "title": "If/Else Switch - Your picture, edits only",
5163 "properties": {
5164 "Node name for S&R": "ComfySwitchNode",
5165 "cnr_id": "comfy-core",
5166 "ver": "0.44.0"
5167 },
5168 "widgets_values": [
5169 true
5170 ],
5171 "widgets_values_named": {
5172 "switch": true
5173 },
5174 "color": "#14424d",
5175 "bgcolor": "#161f21"
5176 },
5177 {
5178 "id": 450,
5179 "type": "AUSBOSS_NODES_MathExpression",
5180 "pos": [
5181 20,
5182 300
5183 ],
5184 "size": [
5185 460,
5186 134
5187 ],
5188 "flags": [],
5189 "order": 22,
5190 "mode": 0,
5191 "inputs": [
5192 {
5193 "localized_name": "a",
5194 "name": "a",
5195 "shape": 7,
5196 "type": "FLOAT,INT",
5197 "link": 1031
5198 },
5199 {
5200 "localized_name": "b",
5201 "name": "b",
5202 "shape": 7,
5203 "type": "FLOAT,INT",
5204 "link": 1032
5205 },
5206 {
5207 "localized_name": "c",
5208 "name": "c",
5209 "shape": 7,
5210 "type": "FLOAT,INT",
5211 "link": null
5212 },
5213 {
5214 "localized_name": "expression",
5215 "name": "expression",
5216 "type": "STRING",
5217 "widget": {
5218 "name": "expression"
5219 },
5220 "link": null
5221 }
5222 ],
5223 "outputs": [
5224 {
5225 "localized_name": "float",
5226 "name": "float",
5227 "type": "FLOAT",
5228 "links": [
5229 1073
5230 ]
5231 },
5232 {
5233 "localized_name": "int",
5234 "name": "int",
5235 "type": "INT",
5236 "links": []
5237 }
5238 ],
5239 "title": "Math Expression - Picture aspect ratio",
5240 "properties": {
5241 "Node name for S&R": "AUSBOSS_NODES_MathExpression",
5242 "cnr_id": "ausboss-nodes",
5243 "ver": "2.15.0"
5244 },
5245 "widgets_values": [
5246 "a / max(b, 1)"
5247 ],
5248 "widgets_values_named": {
5249 "expression": "a / max(b, 1)"
5250 },
5251 "color": "#14424d",
5252 "bgcolor": "#161f21"
5253 },
5254 {
5255 "id": 416,
5256 "type": "ComfySwitchNode",
5257 "pos": [
5258 1100,
5259 1340
5260 ],
5261 "size": [
5262 460,
5263 78
5264 ],
5265 "flags": [],
5266 "order": 13,
5267 "mode": 0,
5268 "inputs": [
5269 {
5270 "localized_name": "on_false",
5271 "name": "on_false",
5272 "shape": 7,
5273 "type": "CLIP",
5274 "link": null
5275 },
5276 {
5277 "localized_name": "on_true",
5278 "name": "on_true",
5279 "shape": 7,
5280 "type": "CLIP",
5281 "link": 1142
5282 },
5283 {
5284 "localized_name": "switch",
5285 "name": "switch",
5286 "type": "BOOLEAN",
5287 "widget": {
5288 "name": "switch"
5289 },
5290 "link": null
5291 }
5292 ],
5293 "outputs": [
5294 {
5295 "localized_name": "output",
5296 "name": "output",
5297 "type": "CLIP",
5298 "links": [
5299 1048,
5300 1049
5301 ]
5302 }
5303 ],
5304 "title": "If/Else Switch - The writer's model, Auto prompt only",
5305 "properties": {
5306 "Node name for S&R": "ComfySwitchNode",
5307 "cnr_id": "comfy-core",
5308 "ver": "0.44.0"
5309 },
5310 "widgets_values": [
5311 true
5312 ],
5313 "widgets_values_named": {
5314 "switch": true
5315 },
5316 "color": "#14424d",
5317 "bgcolor": "#161f21"
5318 },
5319 {
5320 "id": 415,
5321 "type": "ComfySwitchNode",
5322 "pos": [
5323 1100,
5324 1190
5325 ],
5326 "size": [
5327 460,
5328 78
5329 ],
5330 "flags": [],
5331 "order": 13,
5332 "mode": 0,
5333 "inputs": [
5334 {
5335 "localized_name": "on_false",
5336 "name": "on_false",
5337 "shape": 7,
5338 "type": "STRING",
5339 "link": 1144
5340 },
5341 {
5342 "localized_name": "on_true",
5343 "name": "on_true",
5344 "shape": 7,
5345 "type": "STRING",
5346 "link": 1143
5347 },
5348 {
5349 "localized_name": "switch",
5350 "name": "switch",
5351 "type": "BOOLEAN",
5352 "widget": {
5353 "name": "switch"
5354 },
5355 "link": 1145
5356 }
5357 ],
5358 "outputs": [
5359 {
5360 "localized_name": "output",
5361 "name": "output",
5362 "type": "STRING",
5363 "links": [
5364 1056
5365 ]
5366 }
5367 ],
5368 "title": "If/Else Switch - Auto prompt on: the writer. Off: your words",
5369 "properties": {
5370 "Node name for S&R": "ComfySwitchNode",
5371 "cnr_id": "comfy-core",
5372 "ver": "0.44.0"
5373 },
5374 "widgets_values": [
5375 false
5376 ],
5377 "widgets_values_named": {
5378 "switch": false
5379 },
5380 "color": "#14424d",
5381 "bgcolor": "#161f21"
5382 }
5383 ],
5384 "groups": [
5385 {
5386 "id": 5,
5387 "title": "1 · For a new picture",
5388 "bounding": [
5389 0,
5390 0,
5391 500,
5392 674
5393 ],
5394 "color": "#1d5560",
5395 "flags": {
5396 "pinned": true
5397 }
5398 },
5399 {
5400 "id": 6,
5401 "title": "2 · For an edit",
5402 "bounding": [
5403 540,
5404 0,
5405 500,
5406 1780
5407 ],
5408 "color": "#1d5560",
5409 "flags": {
5410 "pinned": true
5411 }
5412 },
5413 {
5414 "id": 7,
5415 "title": "3 · Write",
5416 "bounding": [
5417 1080,
5418 0,
5419 500,
5420 1138
5421 ],
5422 "color": "#1d5560",
5423 "flags": {
5424 "pinned": true
5425 }
5426 },
5427 {
5428 "id": 10,
5429 "title": "4 · Pass the pictures on",
5430 "bounding": [
5431 1600,
5432 0,
5433 500,
5434 1200
5435 ],
5436 "color": "#1d5560",
5437 "flags": {
5438 "pinned": true
5439 }
5440 }
5441 ],
5442 "links": [
5443 {
5444 "id": 1031,
5445 "origin_id": -10,
5446 "origin_slot": 10,
5447 "target_id": 450,
5448 "target_slot": 0,
5449 "type": "INT"
5450 },
5451 {
5452 "id": 1032,
5453 "origin_id": -10,
5454 "origin_slot": 11,
5455 "target_id": 450,
5456 "target_slot": 1,
5457 "type": "INT"
5458 },
5459 {
5460 "id": 1034,
5461 "origin_id": -10,
5462 "origin_slot": 8,
5463 "target_id": 414,
5464 "target_slot": 1,
5465 "type": "IMAGE"
5466 },
5467 {
5468 "id": 1043,
5469 "origin_id": 405,
5470 "origin_slot": 0,
5471 "target_id": 406,
5472 "target_slot": 0,
5473 "type": "IMAGE"
5474 },
5475 {
5476 "id": 1044,
5477 "origin_id": 406,
5478 "origin_slot": 2,
5479 "target_id": 407,
5480 "target_slot": 0,
5481 "type": "INT"
5482 },
5483 {
5484 "id": 1053,
5485 "origin_id": 411,
5486 "origin_slot": 0,
5487 "target_id": 413,
5488 "target_slot": 0,
5489 "type": "STRING"
5490 },
5491 {
5492 "id": 1054,
5493 "origin_id": 412,
5494 "origin_slot": 0,
5495 "target_id": 413,
5496 "target_slot": 1,
5497 "type": "STRING"
5498 },
5499 {
5500 "id": 1055,
5501 "origin_id": -10,
5502 "origin_slot": 12,
5503 "target_id": 413,
5504 "target_slot": 2,
5505 "type": "BOOLEAN"
5506 },
5507 {
5508 "id": 1056,
5509 "origin_id": 415,
5510 "origin_slot": 0,
5511 "target_id": -20,
5512 "target_slot": 0,
5513 "type": "STRING"
5514 },
5515 {
5516 "id": 1057,
5517 "origin_id": -10,
5518 "origin_slot": 1,
5519 "target_id": 421,
5520 "target_slot": 1,
5521 "type": "IMAGE"
5522 },
5523 {
5524 "id": 1058,
5525 "origin_id": 421,
5526 "origin_slot": 0,
5527 "target_id": -20,
5528 "target_slot": 1,
5529 "type": "IMAGE"
5530 },
5531 {
5532 "id": 1059,
5533 "origin_id": -10,
5534 "origin_slot": 2,
5535 "target_id": 422,
5536 "target_slot": 1,
5537 "type": "IMAGE"
5538 },
5539 {
5540 "id": 1060,
5541 "origin_id": 422,
5542 "origin_slot": 0,
5543 "target_id": -20,
5544 "target_slot": 2,
5545 "type": "IMAGE"
5546 },
5547 {
5548 "id": 1061,
5549 "origin_id": -10,
5550 "origin_slot": 3,
5551 "target_id": 423,
5552 "target_slot": 1,
5553 "type": "IMAGE"
5554 },
5555 {
5556 "id": 1062,
5557 "origin_id": 423,
5558 "origin_slot": 0,
5559 "target_id": -20,
5560 "target_slot": 3,
5561 "type": "IMAGE"
5562 },
5563 {
5564 "id": 1063,
5565 "origin_id": -10,
5566 "origin_slot": 4,
5567 "target_id": 424,
5568 "target_slot": 1,
5569 "type": "IMAGE"
5570 },
5571 {
5572 "id": 1064,
5573 "origin_id": 424,
5574 "origin_slot": 0,
5575 "target_id": -20,
5576 "target_slot": 4,
5577 "type": "IMAGE"
5578 },
5579 {
5580 "id": 1065,
5581 "origin_id": -10,
5582 "origin_slot": 5,
5583 "target_id": 425,
5584 "target_slot": 1,
5585 "type": "IMAGE"
5586 },
5587 {
5588 "id": 1066,
5589 "origin_id": 425,
5590 "origin_slot": 0,
5591 "target_id": -20,
5592 "target_slot": 5,
5593 "type": "IMAGE"
5594 },
5595 {
5596 "id": 1067,
5597 "origin_id": -10,
5598 "origin_slot": 6,
5599 "target_id": 426,
5600 "target_slot": 1,
5601 "type": "IMAGE"
5602 },
5603 {
5604 "id": 1068,
5605 "origin_id": 426,
5606 "origin_slot": 0,
5607 "target_id": -20,
5608 "target_slot": 6,
5609 "type": "IMAGE"
5610 },
5611 {
5612 "id": 1069,
5613 "origin_id": -10,
5614 "origin_slot": 7,
5615 "target_id": 427,
5616 "target_slot": 1,
5617 "type": "IMAGE"
5618 },
5619 {
5620 "id": 1070,
5621 "origin_id": 427,
5622 "origin_slot": 0,
5623 "target_id": -20,
5624 "target_slot": 7,
5625 "type": "IMAGE"
5626 },
5627 {
5628 "id": 1071,
5629 "origin_id": -10,
5630 "origin_slot": 12,
5631 "target_id": 414,
5632 "target_slot": 2,
5633 "type": "BOOLEAN"
5634 },
5635 {
5636 "id": 1072,
5637 "origin_id": 414,
5638 "origin_slot": 0,
5639 "target_id": 404,
5640 "target_slot": 0,
5641 "type": "IMAGE"
5642 },
5643 {
5644 "id": 1073,
5645 "origin_id": 450,
5646 "origin_slot": 0,
5647 "target_id": 402,
5648 "target_slot": 0,
5649 "type": "FLOAT"
5650 },
5651 {
5652 "id": 1033,
5653 "origin_id": -10,
5654 "origin_slot": 0,
5655 "target_id": 402,
5656 "target_slot": 3,
5657 "type": "*"
5658 },
5659 {
5660 "id": 1035,
5661 "origin_id": 404,
5662 "origin_slot": 0,
5663 "target_id": 405,
5664 "target_slot": 0,
5665 "type": "IMAGE"
5666 },
5667 {
5668 "id": 1036,
5669 "origin_id": -10,
5670 "origin_slot": 1,
5671 "target_id": 405,
5672 "target_slot": 1,
5673 "type": "IMAGE"
5674 },
5675 {
5676 "id": 1037,
5677 "origin_id": -10,
5678 "origin_slot": 2,
5679 "target_id": 405,
5680 "target_slot": 2,
5681 "type": "IMAGE"
5682 },
5683 {
5684 "id": 1038,
5685 "origin_id": -10,
5686 "origin_slot": 3,
5687 "target_id": 405,
5688 "target_slot": 3,
5689 "type": "IMAGE"
5690 },
5691 {
5692 "id": 1039,
5693 "origin_id": -10,
5694 "origin_slot": 4,
5695 "target_id": 405,
5696 "target_slot": 4,
5697 "type": "IMAGE"
5698 },
5699 {
5700 "id": 1040,
5701 "origin_id": -10,
5702 "origin_slot": 5,
5703 "target_id": 405,
5704 "target_slot": 5,
5705 "type": "IMAGE"
5706 },
5707 {
5708 "id": 1041,
5709 "origin_id": -10,
5710 "origin_slot": 6,
5711 "target_id": 405,
5712 "target_slot": 6,
5713 "type": "IMAGE"
5714 },
5715 {
5716 "id": 1042,
5717 "origin_id": -10,
5718 "origin_slot": 7,
5719 "target_id": 405,
5720 "target_slot": 7,
5721 "type": "IMAGE"
5722 },
5723 {
5724 "id": 1045,
5725 "origin_id": 408,
5726 "origin_slot": 0,
5727 "target_id": 409,
5728 "target_slot": 0,
5729 "type": "STRING"
5730 },
5731 {
5732 "id": 1046,
5733 "origin_id": 407,
5734 "origin_slot": 1,
5735 "target_id": 409,
5736 "target_slot": 1,
5737 "type": "INT"
5738 },
5739 {
5740 "id": 1047,
5741 "origin_id": -10,
5742 "origin_slot": 0,
5743 "target_id": 409,
5744 "target_slot": 3,
5745 "type": "*"
5746 },
5747 {
5748 "id": 1048,
5749 "origin_id": 416,
5750 "origin_slot": 0,
5751 "target_id": 411,
5752 "target_slot": 0,
5753 "type": "CLIP"
5754 },
5755 {
5756 "id": 1050,
5757 "origin_id": 402,
5758 "origin_slot": 0,
5759 "target_id": 411,
5760 "target_slot": 5,
5761 "type": "STRING"
5762 },
5763 {
5764 "id": 1049,
5765 "origin_id": 416,
5766 "origin_slot": 0,
5767 "target_id": 412,
5768 "target_slot": 0,
5769 "type": "CLIP"
5770 },
5771 {
5772 "id": 1052,
5773 "origin_id": 405,
5774 "origin_slot": 0,
5775 "target_id": 412,
5776 "target_slot": 1,
5777 "type": "IMAGE"
5778 },
5779 {
5780 "id": 1051,
5781 "origin_id": 409,
5782 "origin_slot": 0,
5783 "target_id": 412,
5784 "target_slot": 5,
5785 "type": "STRING"
5786 },
5787 {
5788 "id": 1141,
5789 "origin_id": 406,
5790 "origin_slot": 2,
5791 "target_id": 409,
5792 "target_slot": 4,
5793 "type": "INT"
5794 },
5795 {
5796 "id": 1142,
5797 "origin_id": -10,
5798 "origin_slot": 9,
5799 "target_id": 416,
5800 "target_slot": 1,
5801 "type": "CLIP"
5802 },
5803 {
5804 "id": 1143,
5805 "origin_id": 413,
5806 "origin_slot": 0,
5807 "target_id": 415,
5808 "target_slot": 1,
5809 "type": "STRING"
5810 },
5811 {
5812 "id": 1144,
5813 "origin_id": -10,
5814 "origin_slot": 0,
5815 "target_id": 415,
5816 "target_slot": 0,
5817 "type": "STRING"
5818 },
5819 {
5820 "id": 1145,
5821 "origin_id": -10,
5822 "origin_slot": 13,
5823 "target_id": 415,
5824 "target_slot": 2,
5825 "type": "BOOLEAN"
5826 }
5827 ],
5828 "extra": []
5829 },
5830 {
5831 "id": "cb4401eb-1636-558a-bc9a-95a42a91ff55",
5832 "version": 1,
5833 "state": {
5834 "lastGroupId": 22,
5835 "lastNodeId": 470,
5836 "lastLinkId": 1132,
5837 "lastRerouteId": 0
5838 },
5839 "revision": 0,
5840 "config": [],
5841 "name": "Generate or edit with MiniMax-H3-Image",
5842 "inputNode": {
5843 "id": -10,
5844 "bounding": [
5845 -260,
5846 120,
5847 150,
5848 380
5849 ]
5850 },
5851 "outputNode": {
5852 "id": -20,
5853 "bounding": [
5854 1150,
5855 120,
5856 150,
5857 60
5858 ]
5859 },
5860 "inputs": [
5861 {
5862 "id": "e378d806-6279-58a5-a2cb-08e7fa4011d4",
5863 "name": "prompt",
5864 "type": "*",
5865 "linkIds": [
5866 1090
5867 ],
5868 "localized_name": "prompt",
5869 "pos": [
5870 51,
5871 24
5872 ]
5873 },
5874 {
5875 "id": "8c4e36a5-7564-5e38-b016-beb87472a0e9",
5876 "name": "seed",
5877 "type": "*",
5878 "linkIds": [
5879 1096
5880 ],
5881 "localized_name": "seed",
5882 "pos": [
5883 51,
5884 44
5885 ]
5886 },
5887 {
5888 "id": "338a6e40-a095-5c3e-af23-9a33fc6792e9",
5889 "name": "your picture",
5890 "type": "IMAGE",
5891 "linkIds": [
5892 1082,
5893 1106,
5894 1133
5895 ],
5896 "localized_name": "your picture",
5897 "shape": 7,
5898 "pos": [
5899 51,
5900 64
5901 ]
5902 },
5903 {
5904 "id": "dec178a0-9a53-5948-aae6-64bb3e20b5d2",
5905 "name": "width",
5906 "type": "INT",
5907 "linkIds": [
5908 1076
5909 ],
5910 "localized_name": "width",
5911 "shape": 7,
5912 "pos": [
5913 51,
5914 84
5915 ]
5916 },
5917 {
5918 "id": "68f88202-7207-5354-bab8-699d0afb3213",
5919 "name": "height",
5920 "type": "INT",
5921 "linkIds": [
5922 1077
5923 ],
5924 "localized_name": "height",
5925 "shape": 7,
5926 "pos": [
5927 51,
5928 104
5929 ]
5930 },
5931 {
5932 "id": "f4d10fe4-3ea5-5c32-ab52-2f1308cea1b1",
5933 "name": "text encoder",
5934 "type": "CLIP",
5935 "linkIds": [
5936 1080
5937 ],
5938 "localized_name": "text encoder",
5939 "pos": [
5940 51,
5941 124
5942 ]
5943 },
5944 {
5945 "id": "561b9bf2-8043-5d2a-948c-7b754c4ec4f6",
5946 "name": "vae",
5947 "type": "VAE",
5948 "linkIds": [
5949 1081,
5950 1103
5951 ],
5952 "localized_name": "vae",
5953 "pos": [
5954 51,
5955 144
5956 ]
5957 },
5958 {
5959 "id": "73bed593-2556-5fe5-b02a-5701e08a112e",
5960 "name": "diffusion model",
5961 "type": "MODEL",
5962 "linkIds": [
5963 1093
5964 ],
5965 "localized_name": "diffusion model",
5966 "pos": [
5967 51,
5968 164
5969 ]
5970 },
5971 {
5972 "id": "77008f68-bf16-54e5-a7a7-2a53119ec07a",
5973 "name": "picture 2",
5974 "type": "IMAGE",
5975 "linkIds": [
5976 1083
5977 ],
5978 "localized_name": "picture 2",
5979 "shape": 7,
5980 "pos": [
5981 51,
5982 184
5983 ]
5984 },
5985 {
5986 "id": "cec05a3f-bc55-5a4c-9fe7-e65cd65f58f5",
5987 "name": "picture 3",
5988 "type": "IMAGE",
5989 "linkIds": [
5990 1084
5991 ],
5992 "localized_name": "picture 3",
5993 "shape": 7,
5994 "pos": [
5995 51,
5996 204
5997 ]
5998 },
5999 {
6000 "id": "0f2e346e-7dcd-5262-9896-c4d4270482e8",
6001 "name": "picture 4",
6002 "type": "IMAGE",
6003 "linkIds": [
6004 1085
6005 ],
6006 "localized_name": "picture 4",
6007 "shape": 7,
6008 "pos": [
6009 51,
6010 224
6011 ]
6012 },
6013 {
6014 "id": "08177e2f-6189-56f0-ada0-913b1632a0f7",
6015 "name": "picture 5",
6016 "type": "IMAGE",
6017 "linkIds": [
6018 1086
6019 ],
6020 "localized_name": "picture 5",
6021 "shape": 7,
6022 "pos": [
6023 51,
6024 244
6025 ]
6026 },
6027 {
6028 "id": "f7ea84c3-bb52-5564-894e-1be0db999049",
6029 "name": "picture 6",
6030 "type": "IMAGE",
6031 "linkIds": [
6032 1087
6033 ],
6034 "localized_name": "picture 6",
6035 "shape": 7,
6036 "pos": [
6037 51,
6038 264
6039 ]
6040 },
6041 {
6042 "id": "b9e7d92d-563e-54d5-b0f2-84f777da4993",
6043 "name": "picture 7",
6044 "type": "IMAGE",
6045 "linkIds": [
6046 1088
6047 ],
6048 "localized_name": "picture 7",
6049 "shape": 7,
6050 "pos": [
6051 51,
6052 284
6053 ]
6054 },
6055 {
6056 "id": "1c053337-2751-50a5-aaa4-bde5cf99e025",
6057 "name": "picture 8",
6058 "type": "IMAGE",
6059 "linkIds": [
6060 1089
6061 ],
6062 "localized_name": "picture 8",
6063 "shape": 7,
6064 "pos": [
6065 51,
6066 304
6067 ]
6068 },
6069 {
6070 "id": "7de47391-7e2b-5bd5-989a-f4416930bd90",
6071 "name": "size from your picture",
6072 "type": "BOOLEAN",
6073 "linkIds": [
6074 1078,
6075 1079,
6076 1107,
6077 1134
6078 ],
6079 "localized_name": "size from your picture",
6080 "pos": [
6081 51,
6082 324
6083 ]
6084 },
6085 {
6086 "id": "8e51340b-35d0-4f15-8e0a-46661d669a14",
6087 "name": "fast",
6088 "type": "BOOLEAN",
6089 "linkIds": [
6090 1140
6091 ],
6092 "localized_name": "fast",
6093 "pos": [
6094 51,
6095 344
6096 ]
6097 }
6098 ],
6099 "outputs": [
6100 {
6101 "id": "6e33db89-443a-5e03-a35d-16fea34b73fb",
6102 "name": "picture",
6103 "type": "IMAGE",
6104 "linkIds": [
6105 1104
6106 ],
6107 "localized_name": "picture",
6108 "pos": [
6109 24,
6110 24
6111 ]
6112 },
6113 {
6114 "id": "6898cc09-ae31-5e5a-819f-d9fced960b62",
6115 "name": "before",
6116 "type": "IMAGE",
6117 "linkIds": [
6118 1108
6119 ],
6120 "localized_name": "before",
6121 "pos": [
6122 24,
6123 44
6124 ]
6125 }
6126 ],
6127 "widgets": [],
6128 "nodes": [
6129 {
6130 "id": 201,
6131 "type": "AUSBOSS_NODES_Callout",
6132 "pos": [
6133 20,
6134 80
6135 ],
6136 "size": [
6137 460,
6138 178
6139 ],
6140 "flags": {
6141 "pinned": true
6142 },
6143 "order": 0,
6144 "mode": 0,
6145 "inputs": [
6146 {
6147 "localized_name": "text",
6148 "name": "text",
6149 "type": "STRING",
6150 "widget": {
6151 "name": "text"
6152 },
6153 "link": null
6154 }
6155 ],
6156 "outputs": [],
6157 "title": "Callout",
6158 "properties": {
6159 "Node name for S&R": "AUSBOSS_NODES_Callout",
6160 "cnr_id": "ausboss-nodes"
6161 },
6162 "widgets_values": [
6163 "MiniMax-H3-Image reads your prompt, your picture as Picture 1 and any extra pictures as Picture 2 and up.\n###### For an edit the size is your picture's, already scaled to 4 or 16 MP. For a new picture it comes from Latent Size."
6164 ],
6165 "widgets_values_named": {
6166 "text": "MiniMax-H3-Image reads your prompt, your picture as Picture 1 and any extra pictures as Picture 2 and up.\n###### For an edit the size is your picture's, already scaled to 4 or 16 MP. For a new picture it comes from Latent Size."
6167 },
6168 "color": "#07090a",
6169 "bgcolor": "#07090a"
6170 },
6171 {
6172 "id": 203,
6173 "type": "ComfySwitchNode",
6174 "pos": [
6175 20,
6176 492
6177 ],
6178 "size": [
6179 460,
6180 78
6181 ],
6182 "flags": [],
6183 "order": 5,
6184 "mode": 0,
6185 "inputs": [
6186 {
6187 "localized_name": "on_false",
6188 "name": "on_false",
6189 "shape": 7,
6190 "type": "INT",
6191 "link": 1076
6192 },
6193 {
6194 "localized_name": "on_true",
6195 "name": "on_true",
6196 "shape": 7,
6197 "type": "INT",
6198 "link": 1074
6199 },
6200 {
6201 "localized_name": "switch",
6202 "name": "switch",
6203 "type": "BOOLEAN",
6204 "widget": {
6205 "name": "switch"
6206 },
6207 "link": 1078
6208 }
6209 ],
6210 "outputs": [
6211 {
6212 "localized_name": "output",
6213 "name": "output",
6214 "type": "INT",
6215 "links": [
6216 1091
6217 ]
6218 }
6219 ],
6220 "title": "If/Else Switch - Whose width",
6221 "properties": {
6222 "Node name for S&R": "ComfySwitchNode",
6223 "cnr_id": "comfy-core",
6224 "ver": "0.44.0"
6225 },
6226 "widgets_values": [
6227 true
6228 ],
6229 "widgets_values_named": {
6230 "switch": true
6231 },
6232 "color": "#14424d",
6233 "bgcolor": "#161f21"
6234 },
6235 {
6236 "id": 204,
6237 "type": "ComfySwitchNode",
6238 "pos": [
6239 20,
6240 640
6241 ],
6242 "size": [
6243 460,
6244 78
6245 ],
6246 "flags": [],
6247 "order": 6,
6248 "mode": 0,
6249 "inputs": [
6250 {
6251 "localized_name": "on_false",
6252 "name": "on_false",
6253 "shape": 7,
6254 "type": "INT",
6255 "link": 1077
6256 },
6257 {
6258 "localized_name": "on_true",
6259 "name": "on_true",
6260 "shape": 7,
6261 "type": "INT",
6262 "link": 1075
6263 },
6264 {
6265 "localized_name": "switch",
6266 "name": "switch",
6267 "type": "BOOLEAN",
6268 "widget": {
6269 "name": "switch"
6270 },
6271 "link": 1079
6272 }
6273 ],
6274 "outputs": [
6275 {
6276 "localized_name": "output",
6277 "name": "output",
6278 "type": "INT",
6279 "links": [
6280 1092
6281 ]
6282 }
6283 ],
6284 "title": "If/Else Switch - Whose height",
6285 "properties": {
6286 "Node name for S&R": "ComfySwitchNode",
6287 "cnr_id": "comfy-core",
6288 "ver": "0.44.0"
6289 },
6290 "widgets_values": [
6291 true
6292 ],
6293 "widgets_values_named": {
6294 "switch": true
6295 },
6296 "color": "#14424d",
6297 "bgcolor": "#161f21"
6298 },
6299 {
6300 "id": 205,
6301 "type": "MiniMaxH3ImageConditioning",
6302 "pos": [
6303 20,
6304 788
6305 ],
6306 "size": [
6307 460,
6308 356
6309 ],
6310 "flags": [],
6311 "order": 7,
6312 "mode": 0,
6313 "inputs": [
6314 {
6315 "localized_name": "clip",
6316 "name": "clip",
6317 "type": "CLIP",
6318 "link": 1080
6319 },
6320 {
6321 "localized_name": "vae",
6322 "name": "vae",
6323 "shape": 7,
6324 "type": "VAE",
6325 "link": 1081
6326 },
6327 {
6328 "localized_name": "image_1",
6329 "name": "images.image_1",
6330 "shape": 7,
6331 "type": "IMAGE",
6332 "link": 1082
6333 },
6334 {
6335 "localized_name": "image_2",
6336 "name": "images.image_2",
6337 "shape": 7,
6338 "type": "IMAGE",
6339 "link": 1083
6340 },
6341 {
6342 "localized_name": "image_3",
6343 "name": "images.image_3",
6344 "shape": 7,
6345 "type": "IMAGE",
6346 "link": 1084
6347 },
6348 {
6349 "localized_name": "image_4",
6350 "name": "images.image_4",
6351 "shape": 7,
6352 "type": "IMAGE",
6353 "link": 1085
6354 },
6355 {
6356 "localized_name": "image_5",
6357 "name": "images.image_5",
6358 "shape": 7,
6359 "type": "IMAGE",
6360 "link": 1086
6361 },
6362 {
6363 "localized_name": "image_6",
6364 "name": "images.image_6",
6365 "shape": 7,
6366 "type": "IMAGE",
6367 "link": 1087
6368 },
6369 {
6370 "localized_name": "image_7",
6371 "name": "images.image_7",
6372 "shape": 7,
6373 "type": "IMAGE",
6374 "link": 1088
6375 },
6376 {
6377 "localized_name": "image_8",
6378 "name": "images.image_8",
6379 "shape": 7,
6380 "type": "IMAGE",
6381 "link": 1089
6382 },
6383 {
6384 "localized_name": "prompt",
6385 "name": "prompt",
6386 "type": "STRING",
6387 "widget": {
6388 "name": "prompt"
6389 },
6390 "link": 1090
6391 },
6392 {
6393 "localized_name": "width",
6394 "name": "width",
6395 "type": "INT",
6396 "widget": {
6397 "name": "width"
6398 },
6399 "link": 1091
6400 },
6401 {
6402 "localized_name": "height",
6403 "name": "height",
6404 "type": "INT",
6405 "widget": {
6406 "name": "height"
6407 },
6408 "link": 1092
6409 }
6410 ],
6411 "outputs": [
6412 {
6413 "localized_name": "positive",
6414 "name": "positive",
6415 "type": "CONDITIONING",
6416 "links": [
6417 1095
6418 ]
6419 },
6420 {
6421 "localized_name": "LATENT",
6422 "name": "LATENT",
6423 "type": "LATENT",
6424 "links": [
6425 1101
6426 ]
6427 }
6428 ],
6429 "title": "MiniMax H3 Image Conditioning - Your prompt and pictures",
6430 "properties": {
6431 "Node name for S&R": "MiniMaxH3ImageConditioning",
6432 "cnr_id": "comfy-core",
6433 "ver": "ede4e5fe8"
6434 },
6435 "widgets_values": [
6436 "",
6437 0,
6438 0
6439 ],
6440 "widgets_values_named": {
6441 "prompt": "",
6442 "width": 0,
6443 "height": 0
6444 },
6445 "color": "#14424d",
6446 "bgcolor": "#161f21"
6447 },
6448 {
6449 "id": 206,
6450 "type": "AUSBOSS_NODES_Callout",
6451 "pos": [
6452 560,
6453 80
6454 ],
6455 "size": [
6456 460,
6457 178
6458 ],
6459 "flags": {
6460 "pinned": true
6461 },
6462 "order": 2,
6463 "mode": 0,
6464 "inputs": [
6465 {
6466 "localized_name": "text",
6467 "name": "text",
6468 "type": "STRING",
6469 "widget": {
6470 "name": "text"
6471 },
6472 "link": null
6473 }
6474 ],
6475 "outputs": [],
6476 "title": "Callout",
6477 "properties": {
6478 "Node name for S&R": "AUSBOSS_NODES_Callout",
6479 "cnr_id": "ausboss-nodes"
6480 },
6481 "widgets_values": [
6482 "6 steps, no CFG: the model is made for it. **Fast** swaps in the pDMD LoRA's 2 steps.\n###### The maker's own 6-step list and euler. No shift node: the model keeps its own schedule. The seed comes from the top level."
6483 ],
6484 "widgets_values_named": {
6485 "text": "6 steps, no CFG: the model is made for it. **Fast** swaps in the pDMD LoRA's 2 steps.\n###### The maker's own 6-step list and euler. No shift node: the model keeps its own schedule. The seed comes from the top level."
6486 },
6487 "color": "#07090a",
6488 "bgcolor": "#07090a"
6489 },
6490 {
6491 "id": 208,
6492 "type": "BasicGuider",
6493 "pos": [
6494 560,
6495 508
6496 ],
6497 "size": [
6498 460,
6499 50
6500 ],
6501 "flags": [],
6502 "order": 9,
6503 "mode": 0,
6504 "inputs": [
6505 {
6506 "localized_name": "model",
6507 "name": "model",
6508 "type": "MODEL",
6509 "link": 1093
6510 },
6511 {
6512 "localized_name": "conditioning",
6513 "name": "conditioning",
6514 "type": "CONDITIONING",
6515 "link": 1095
6516 }
6517 ],
6518 "outputs": [
6519 {
6520 "localized_name": "GUIDER",
6521 "name": "GUIDER",
6522 "type": "GUIDER",
6523 "links": [
6524 1098
6525 ]
6526 }
6527 ],
6528 "title": "Basic Guider - No CFG",
6529 "properties": {
6530 "Node name for S&R": "BasicGuider",
6531 "cnr_id": "comfy-core",
6532 "ver": "0.44.0"
6533 },
6534 "color": "#14424d",
6535 "bgcolor": "#161f21"
6536 },
6537 {
6538 "id": 209,
6539 "type": "RandomNoise",
6540 "pos": [
6541 560,
6542 628
6543 ],
6544 "size": [
6545 460,
6546 82
6547 ],
6548 "flags": [],
6549 "order": 10,
6550 "mode": 0,
6551 "inputs": [
6552 {
6553 "localized_name": "noise_seed",
6554 "name": "noise_seed",
6555 "type": "INT",
6556 "widget": {
6557 "name": "noise_seed"
6558 },
6559 "link": 1096
6560 }
6561 ],
6562 "outputs": [
6563 {
6564 "localized_name": "NOISE",
6565 "name": "NOISE",
6566 "type": "NOISE",
6567 "links": [
6568 1097
6569 ]
6570 }
6571 ],
6572 "title": "Random Noise - Seed from the top level",
6573 "properties": {
6574 "Node name for S&R": "RandomNoise",
6575 "cnr_id": "comfy-core",
6576 "ver": "0.44.0"
6577 },
6578 "widgets_values": [
6579 0,
6580 "fixed"
6581 ],
6582 "widgets_values_named": {
6583 "noise_seed": 0,
6584 "control_after_generate": "fixed"
6585 },
6586 "color": "#14424d",
6587 "bgcolor": "#161f21"
6588 },
6589 {
6590 "id": 210,
6591 "type": "KSamplerSelect",
6592 "pos": [
6593 560,
6594 780
6595 ],
6596 "size": [
6597 460,
6598 58
6599 ],
6600 "flags": [],
6601 "order": 3,
6602 "mode": 0,
6603 "inputs": [
6604 {
6605 "localized_name": "sampler_name",
6606 "name": "sampler_name",
6607 "type": "COMBO",
6608 "widget": {
6609 "name": "sampler_name"
6610 },
6611 "link": null
6612 }
6613 ],
6614 "outputs": [
6615 {
6616 "localized_name": "SAMPLER",
6617 "name": "SAMPLER",
6618 "type": "SAMPLER",
6619 "links": [
6620 1099
6621 ]
6622 }
6623 ],
6624 "title": "Sampler: euler",
6625 "properties": {
6626 "Node name for S&R": "KSamplerSelect",
6627 "cnr_id": "comfy-core",
6628 "ver": "0.44.0"
6629 },
6630 "widgets_values": [
6631 "euler"
6632 ],
6633 "widgets_values_named": {
6634 "sampler_name": "euler"
6635 },
6636 "color": "#14424d",
6637 "bgcolor": "#161f21"
6638 },
6639 {
6640 "id": 211,
6641 "type": "ManualSigmas",
6642 "pos": [
6643 560,
6644 908
6645 ],
6646 "size": [
6647 460,
6648 100
6649 ],
6650 "flags": [],
6651 "order": 4,
6652 "mode": 0,
6653 "inputs": [
6654 {
6655 "localized_name": "sigmas",
6656 "name": "sigmas",
6657 "type": "STRING",
6658 "widget": {
6659 "name": "sigmas"
6660 },
6661 "link": null
6662 }
6663 ],
6664 "outputs": [
6665 {
6666 "localized_name": "SIGMAS",
6667 "name": "SIGMAS",
6668 "type": "SIGMAS",
6669 "links": [
6670 1100
6671 ]
6672 }
6673 ],
6674 "title": "The maker's 6 steps",
6675 "properties": {
6676 "Node name for S&R": "ManualSigmas",
6677 "cnr_id": "comfy-core",
6678 "ver": "0.44.0"
6679 },
6680 "widgets_values": [
6681 "1.0, 0.967524, 0.887549, 0.75, 0.532774, 0.232009, 0.0"
6682 ],
6683 "widgets_values_named": {
6684 "sigmas": "1.0, 0.967524, 0.887549, 0.75, 0.532774, 0.232009, 0.0"
6685 },
6686 "color": "#14424d",
6687 "bgcolor": "#161f21"
6688 },
6689 {
6690 "id": 212,
6691 "type": "SamplerCustomAdvanced",
6692 "pos": [
6693 560,
6694 1078
6695 ],
6696 "size": [
6697 460,
6698 110
6699 ],
6700 "flags": [],
6701 "order": 11,
6702 "mode": 0,
6703 "inputs": [
6704 {
6705 "localized_name": "noise",
6706 "name": "noise",
6707 "type": "NOISE",
6708 "link": 1097
6709 },
6710 {
6711 "localized_name": "guider",
6712 "name": "guider",
6713 "type": "GUIDER",
6714 "link": 1098
6715 },
6716 {
6717 "localized_name": "sampler",
6718 "name": "sampler",
6719 "type": "SAMPLER",
6720 "link": 1099
6721 },
6722 {
6723 "localized_name": "sigmas",
6724 "name": "sigmas",
6725 "type": "SIGMAS",
6726 "link": 1139
6727 },
6728 {
6729 "localized_name": "latent_image",
6730 "name": "latent_image",
6731 "type": "LATENT",
6732 "link": 1101
6733 }
6734 ],
6735 "outputs": [
6736 {
6737 "localized_name": "output",
6738 "name": "output",
6739 "type": "LATENT",
6740 "links": [
6741 1102
6742 ]
6743 },
6744 {
6745 "localized_name": "denoised_output",
6746 "name": "denoised_output",
6747 "type": "LATENT",
6748 "links": []
6749 }
6750 ],
6751 "title": "Sampler - 6 steps, or 2 with Fast",
6752 "properties": {
6753 "Node name for S&R": "SamplerCustomAdvanced",
6754 "cnr_id": "comfy-core",
6755 "ver": "0.44.0"
6756 },
6757 "color": "#14424d",
6758 "bgcolor": "#161f21"
6759 },
6760 {
6761 "id": 213,
6762 "type": "VAEDecode",
6763 "pos": [
6764 560,
6765 1258
6766 ],
6767 "size": [
6768 460,
6769 46
6770 ],
6771 "flags": [],
6772 "order": 12,
6773 "mode": 0,
6774 "inputs": [
6775 {
6776 "localized_name": "samples",
6777 "name": "samples",
6778 "type": "LATENT",
6779 "link": 1102
6780 },
6781 {
6782 "localized_name": "vae",
6783 "name": "vae",
6784 "type": "VAE",
6785 "link": 1103
6786 }
6787 ],
6788 "outputs": [
6789 {
6790 "localized_name": "IMAGE",
6791 "name": "IMAGE",
6792 "type": "IMAGE",
6793 "links": [
6794 1104,
6795 1105
6796 ]
6797 }
6798 ],
6799 "title": "Decode",
6800 "properties": {
6801 "Node name for S&R": "VAEDecode",
6802 "cnr_id": "comfy-core",
6803 "ver": "0.44.0"
6804 },
6805 "color": "#14424d",
6806 "bgcolor": "#161f21"
6807 },
6808 {
6809 "id": 460,
6810 "type": "ComfySwitchNode",
6811 "pos": [
6812 560,
6813 1374
6814 ],
6815 "size": [
6816 460,
6817 78
6818 ],
6819 "flags": [],
6820 "order": 13,
6821 "mode": 0,
6822 "inputs": [
6823 {
6824 "localized_name": "on_false",
6825 "name": "on_false",
6826 "shape": 7,
6827 "type": "IMAGE",
6828 "link": 1105
6829 },
6830 {
6831 "localized_name": "on_true",
6832 "name": "on_true",
6833 "shape": 7,
6834 "type": "IMAGE",
6835 "link": 1106
6836 },
6837 {
6838 "localized_name": "switch",
6839 "name": "switch",
6840 "type": "BOOLEAN",
6841 "widget": {
6842 "name": "switch"
6843 },
6844 "link": 1107
6845 }
6846 ],
6847 "outputs": [
6848 {
6849 "localized_name": "output",
6850 "name": "output",
6851 "type": "IMAGE",
6852 "links": [
6853 1108
6854 ]
6855 }
6856 ],
6857 "title": "If/Else Switch - Original for an edit, result for a new picture",
6858 "properties": {
6859 "Node name for S&R": "ComfySwitchNode",
6860 "cnr_id": "comfy-core",
6861 "ver": "0.44.0"
6862 },
6863 "widgets_values": [
6864 true
6865 ],
6866 "widgets_values_named": {
6867 "switch": true
6868 },
6869 "color": "#14424d",
6870 "bgcolor": "#161f21"
6871 },
6872 {
6873 "id": 473,
6874 "type": "ComfySwitchNode",
6875 "pos": [
6876 -240,
6877 328
6878 ],
6879 "size": [
6880 460,
6881 78
6882 ],
6883 "flags": [],
6884 "order": 13,
6885 "mode": 0,
6886 "inputs": [
6887 {
6888 "localized_name": "on_false",
6889 "name": "on_false",
6890 "shape": 7,
6891 "type": "IMAGE",
6892 "link": null
6893 },
6894 {
6895 "localized_name": "on_true",
6896 "name": "on_true",
6897 "shape": 7,
6898 "type": "IMAGE",
6899 "link": 1133
6900 },
6901 {
6902 "localized_name": "switch",
6903 "name": "switch",
6904 "type": "BOOLEAN",
6905 "widget": {
6906 "name": "switch"
6907 },
6908 "link": 1134
6909 }
6910 ],
6911 "outputs": [
6912 {
6913 "localized_name": "output",
6914 "name": "output",
6915 "type": "IMAGE",
6916 "links": [
6917 1135
6918 ]
6919 }
6920 ],
6921 "title": "If/Else Switch - Your picture, edits only",
6922 "properties": {
6923 "Node name for S&R": "ComfySwitchNode",
6924 "cnr_id": "comfy-core",
6925 "ver": "0.44.0"
6926 },
6927 "widgets_values": [
6928 true
6929 ],
6930 "widgets_values_named": {
6931 "switch": true
6932 },
6933 "color": "#14424d",
6934 "bgcolor": "#161f21"
6935 },
6936 {
6937 "id": 470,
6938 "type": "GetImageSize",
6939 "pos": [
6940 20,
6941 328
6942 ],
6943 "size": [
6944 460,
6945 90
6946 ],
6947 "flags": [],
6948 "order": 8,
6949 "mode": 0,
6950 "inputs": [
6951 {
6952 "localized_name": "image",
6953 "name": "image",
6954 "type": "IMAGE",
6955 "link": 1135
6956 }
6957 ],
6958 "outputs": [
6959 {
6960 "localized_name": "width",
6961 "name": "width",
6962 "type": "INT",
6963 "links": [
6964 1136
6965 ]
6966 },
6967 {
6968 "localized_name": "height",
6969 "name": "height",
6970 "type": "INT",
6971 "links": [
6972 1137
6973 ]
6974 },
6975 {
6976 "localized_name": "batch_size",
6977 "name": "batch_size",
6978 "type": "INT",
6979 "links": []
6980 }
6981 ],
6982 "title": "Get Image Size - Your picture",
6983 "properties": {
6984 "Node name for S&R": "GetImageSize",
6985 "cnr_id": "comfy-core",
6986 "ver": "0.44.0"
6987 },
6988 "color": "#14424d",
6989 "bgcolor": "#161f21"
6990 },
6991 {
6992 "id": 471,
6993 "type": "AUSBOSS_NODES_MathExpression",
6994 "pos": [
6995 20,
6996 438
6997 ],
6998 "size": [
6999 300,
7000 134
7001 ],
7002 "flags": {
7003 "collapsed": true
7004 },
7005 "order": 24,
7006 "mode": 0,
7007 "inputs": [
7008 {
7009 "localized_name": "a",
7010 "name": "a",
7011 "shape": 7,
7012 "type": "FLOAT,INT",
7013 "link": 1136
7014 },
7015 {
7016 "localized_name": "b",
7017 "name": "b",
7018 "shape": 7,
7019 "type": "FLOAT,INT",
7020 "link": null
7021 },
7022 {
7023 "localized_name": "c",
7024 "name": "c",
7025 "shape": 7,
7026 "type": "FLOAT,INT",
7027 "link": null
7028 },
7029 {
7030 "localized_name": "expression",
7031 "name": "expression",
7032 "type": "STRING",
7033 "widget": {
7034 "name": "expression"
7035 },
7036 "link": null
7037 }
7038 ],
7039 "outputs": [
7040 {
7041 "localized_name": "float",
7042 "name": "float",
7043 "type": "FLOAT",
7044 "links": []
7045 },
7046 {
7047 "localized_name": "int",
7048 "name": "int",
7049 "type": "INT",
7050 "links": [
7051 1074
7052 ]
7053 }
7054 ],
7055 "title": "Width of your picture, to 32",
7056 "properties": {
7057 "Node name for S&R": "AUSBOSS_NODES_MathExpression"
7058 },
7059 "widgets_values": [
7060 "round(a / 32) * 32"
7061 ],
7062 "widgets_values_named": {
7063 "expression": "round(a / 32) * 32"
7064 },
7065 "color": "#14424d",
7066 "bgcolor": "#161f21"
7067 },
7068 {
7069 "id": 472,
7070 "type": "AUSBOSS_NODES_MathExpression",
7071 "pos": [
7072 20,
7073 548
7074 ],
7075 "size": [
7076 300,
7077 134
7078 ],
7079 "flags": {
7080 "collapsed": true
7081 },
7082 "order": 24,
7083 "mode": 0,
7084 "inputs": [
7085 {
7086 "localized_name": "a",
7087 "name": "a",
7088 "shape": 7,
7089 "type": "FLOAT,INT",
7090 "link": 1137
7091 },
7092 {
7093 "localized_name": "b",
7094 "name": "b",
7095 "shape": 7,
7096 "type": "FLOAT,INT",
7097 "link": null
7098 },
7099 {
7100 "localized_name": "c",
7101 "name": "c",
7102 "shape": 7,
7103 "type": "FLOAT,INT",
7104 "link": null
7105 },
7106 {
7107 "localized_name": "expression",
7108 "name": "expression",
7109 "type": "STRING",
7110 "widget": {
7111 "name": "expression"
7112 },
7113 "link": null
7114 }
7115 ],
7116 "outputs": [
7117 {
7118 "localized_name": "float",
7119 "name": "float",
7120 "type": "FLOAT",
7121 "links": []
7122 },
7123 {
7124 "localized_name": "int",
7125 "name": "int",
7126 "type": "INT",
7127 "links": [
7128 1075
7129 ]
7130 }
7131 ],
7132 "title": "Height of your picture, to 32",
7133 "properties": {
7134 "Node name for S&R": "AUSBOSS_NODES_MathExpression"
7135 },
7136 "widgets_values": [
7137 "round(a / 32) * 32"
7138 ],
7139 "widgets_values_named": {
7140 "expression": "round(a / 32) * 32"
7141 },
7142 "color": "#14424d",
7143 "bgcolor": "#161f21"
7144 },
7145 {
7146 "id": 474,
7147 "type": "ManualSigmas",
7148 "pos": [
7149 560,
7150 1048
7151 ],
7152 "size": [
7153 460,
7154 100
7155 ],
7156 "flags": [],
7157 "order": 4,
7158 "mode": 0,
7159 "inputs": [
7160 {
7161 "localized_name": "sigmas",
7162 "name": "sigmas",
7163 "type": "STRING",
7164 "widget": {
7165 "name": "sigmas"
7166 },
7167 "link": null
7168 }
7169 ],
7170 "outputs": [
7171 {
7172 "localized_name": "SIGMAS",
7173 "name": "SIGMAS",
7174 "type": "SIGMAS",
7175 "links": [
7176 1138
7177 ]
7178 }
7179 ],
7180 "title": "pDMD's 2 steps (Fast)",
7181 "properties": {
7182 "Node name for S&R": "ManualSigmas",
7183 "cnr_id": "comfy-core",
7184 "ver": "0.44.0"
7185 },
7186 "widgets_values": [
7187 "1.0, 0.923077, 0.0"
7188 ],
7189 "widgets_values_named": {
7190 "sigmas": "1.0, 0.923077, 0.0"
7191 },
7192 "color": "#14424d",
7193 "bgcolor": "#161f21"
7194 },
7195 {
7196 "id": 475,
7197 "type": "ComfySwitchNode",
7198 "pos": [
7199 560,
7200 1188
7201 ],
7202 "size": [
7203 460,
7204 78
7205 ],
7206 "flags": [],
7207 "order": 13,
7208 "mode": 0,
7209 "inputs": [
7210 {
7211 "localized_name": "on_false",
7212 "name": "on_false",
7213 "shape": 7,
7214 "type": "SIGMAS",
7215 "link": 1100
7216 },
7217 {
7218 "localized_name": "on_true",
7219 "name": "on_true",
7220 "shape": 7,
7221 "type": "SIGMAS",
7222 "link": 1138
7223 },
7224 {
7225 "localized_name": "switch",
7226 "name": "switch",
7227 "type": "BOOLEAN",
7228 "widget": {
7229 "name": "switch"
7230 },
7231 "link": 1140
7232 }
7233 ],
7234 "outputs": [
7235 {
7236 "localized_name": "output",
7237 "name": "output",
7238 "type": "SIGMAS",
7239 "links": [
7240 1139
7241 ]
7242 }
7243 ],
7244 "title": "If/Else Switch - Fast or the maker's 6 steps",
7245 "properties": {
7246 "Node name for S&R": "ComfySwitchNode",
7247 "cnr_id": "comfy-core",
7248 "ver": "0.44.0"
7249 },
7250 "widgets_values": [
7251 false
7252 ],
7253 "widgets_values_named": {
7254 "switch": false
7255 },
7256 "color": "#14424d",
7257 "bgcolor": "#161f21"
7258 }
7259 ],
7260 "groups": [
7261 {
7262 "id": 8,
7263 "title": "1 · Read",
7264 "bounding": [
7265 0,
7266 0,
7267 500,
7268 1138
7269 ],
7270 "color": "#1d5560",
7271 "flags": {
7272 "pinned": true
7273 }
7274 },
7275 {
7276 "id": 9,
7277 "title": "2 · Draw",
7278 "bounding": [
7279 540,
7280 0,
7281 500,
7282 1472
7283 ],
7284 "color": "#1d5560",
7285 "flags": {
7286 "pinned": true
7287 }
7288 }
7289 ],
7290 "links": [
7291 {
7292 "id": 1074,
7293 "origin_id": 471,
7294 "origin_slot": 1,
7295 "target_id": 203,
7296 "target_slot": 1,
7297 "type": "INT"
7298 },
7299 {
7300 "id": 1075,
7301 "origin_id": 472,
7302 "origin_slot": 1,
7303 "target_id": 204,
7304 "target_slot": 1,
7305 "type": "INT"
7306 },
7307 {
7308 "id": 1076,
7309 "origin_id": -10,
7310 "origin_slot": 3,
7311 "target_id": 203,
7312 "target_slot": 0,
7313 "type": "INT"
7314 },
7315 {
7316 "id": 1077,
7317 "origin_id": -10,
7318 "origin_slot": 4,
7319 "target_id": 204,
7320 "target_slot": 0,
7321 "type": "INT"
7322 },
7323 {
7324 "id": 1078,
7325 "origin_id": -10,
7326 "origin_slot": 15,
7327 "target_id": 203,
7328 "target_slot": 2,
7329 "type": "BOOLEAN"
7330 },
7331 {
7332 "id": 1079,
7333 "origin_id": -10,
7334 "origin_slot": 15,
7335 "target_id": 204,
7336 "target_slot": 2,
7337 "type": "BOOLEAN"
7338 },
7339 {
7340 "id": 1093,
7341 "origin_id": -10,
7342 "origin_slot": 7,
7343 "target_id": 208,
7344 "target_slot": 0,
7345 "type": "MODEL"
7346 },
7347 {
7348 "id": 1095,
7349 "origin_id": 205,
7350 "origin_slot": 0,
7351 "target_id": 208,
7352 "target_slot": 1,
7353 "type": "CONDITIONING"
7354 },
7355 {
7356 "id": 1096,
7357 "origin_id": -10,
7358 "origin_slot": 1,
7359 "target_id": 209,
7360 "target_slot": 0,
7361 "type": "*"
7362 },
7363 {
7364 "id": 1097,
7365 "origin_id": 209,
7366 "origin_slot": 0,
7367 "target_id": 212,
7368 "target_slot": 0,
7369 "type": "NOISE"
7370 },
7371 {
7372 "id": 1098,
7373 "origin_id": 208,
7374 "origin_slot": 0,
7375 "target_id": 212,
7376 "target_slot": 1,
7377 "type": "GUIDER"
7378 },
7379 {
7380 "id": 1099,
7381 "origin_id": 210,
7382 "origin_slot": 0,
7383 "target_id": 212,
7384 "target_slot": 2,
7385 "type": "SAMPLER"
7386 },
7387 {
7388 "id": 1100,
7389 "origin_id": 211,
7390 "origin_slot": 0,
7391 "target_id": 475,
7392 "target_slot": 0,
7393 "type": "SIGMAS"
7394 },
7395 {
7396 "id": 1101,
7397 "origin_id": 205,
7398 "origin_slot": 1,
7399 "target_id": 212,
7400 "target_slot": 4,
7401 "type": "LATENT"
7402 },
7403 {
7404 "id": 1102,
7405 "origin_id": 212,
7406 "origin_slot": 0,
7407 "target_id": 213,
7408 "target_slot": 0,
7409 "type": "LATENT"
7410 },
7411 {
7412 "id": 1103,
7413 "origin_id": -10,
7414 "origin_slot": 6,
7415 "target_id": 213,
7416 "target_slot": 1,
7417 "type": "VAE"
7418 },
7419 {
7420 "id": 1104,
7421 "origin_id": 213,
7422 "origin_slot": 0,
7423 "target_id": -20,
7424 "target_slot": 0,
7425 "type": "IMAGE"
7426 },
7427 {
7428 "id": 1105,
7429 "origin_id": 213,
7430 "origin_slot": 0,
7431 "target_id": 460,
7432 "target_slot": 0,
7433 "type": "IMAGE"
7434 },
7435 {
7436 "id": 1106,
7437 "origin_id": -10,
7438 "origin_slot": 2,
7439 "target_id": 460,
7440 "target_slot": 1,
7441 "type": "IMAGE"
7442 },
7443 {
7444 "id": 1107,
7445 "origin_id": -10,
7446 "origin_slot": 15,
7447 "target_id": 460,
7448 "target_slot": 2,
7449 "type": "BOOLEAN"
7450 },
7451 {
7452 "id": 1108,
7453 "origin_id": 460,
7454 "origin_slot": 0,
7455 "target_id": -20,
7456 "target_slot": 1,
7457 "type": "IMAGE"
7458 },
7459 {
7460 "id": 1080,
7461 "origin_id": -10,
7462 "origin_slot": 5,
7463 "target_id": 205,
7464 "target_slot": 0,
7465 "type": "CLIP"
7466 },
7467 {
7468 "id": 1081,
7469 "origin_id": -10,
7470 "origin_slot": 6,
7471 "target_id": 205,
7472 "target_slot": 1,
7473 "type": "VAE"
7474 },
7475 {
7476 "id": 1082,
7477 "origin_id": -10,
7478 "origin_slot": 2,
7479 "target_id": 205,
7480 "target_slot": 2,
7481 "type": "IMAGE"
7482 },
7483 {
7484 "id": 1083,
7485 "origin_id": -10,
7486 "origin_slot": 8,
7487 "target_id": 205,
7488 "target_slot": 3,
7489 "type": "IMAGE"
7490 },
7491 {
7492 "id": 1084,
7493 "origin_id": -10,
7494 "origin_slot": 9,
7495 "target_id": 205,
7496 "target_slot": 4,
7497 "type": "IMAGE"
7498 },
7499 {
7500 "id": 1085,
7501 "origin_id": -10,
7502 "origin_slot": 10,
7503 "target_id": 205,
7504 "target_slot": 5,
7505 "type": "IMAGE"
7506 },
7507 {
7508 "id": 1086,
7509 "origin_id": -10,
7510 "origin_slot": 11,
7511 "target_id": 205,
7512 "target_slot": 6,
7513 "type": "IMAGE"
7514 },
7515 {
7516 "id": 1087,
7517 "origin_id": -10,
7518 "origin_slot": 12,
7519 "target_id": 205,
7520 "target_slot": 7,
7521 "type": "IMAGE"
7522 },
7523 {
7524 "id": 1088,
7525 "origin_id": -10,
7526 "origin_slot": 13,
7527 "target_id": 205,
7528 "target_slot": 8,
7529 "type": "IMAGE"
7530 },
7531 {
7532 "id": 1089,
7533 "origin_id": -10,
7534 "origin_slot": 14,
7535 "target_id": 205,
7536 "target_slot": 9,
7537 "type": "IMAGE"
7538 },
7539 {
7540 "id": 1090,
7541 "origin_id": -10,
7542 "origin_slot": 0,
7543 "target_id": 205,
7544 "target_slot": 11,
7545 "type": "*"
7546 },
7547 {
7548 "id": 1091,
7549 "origin_id": 203,
7550 "origin_slot": 0,
7551 "target_id": 205,
7552 "target_slot": 12,
7553 "type": "INT"
7554 },
7555 {
7556 "id": 1092,
7557 "origin_id": 204,
7558 "origin_slot": 0,
7559 "target_id": 205,
7560 "target_slot": 13,
7561 "type": "INT"
7562 },
7563 {
7564 "id": 1133,
7565 "origin_id": -10,
7566 "origin_slot": 2,
7567 "target_id": 473,
7568 "target_slot": 1,
7569 "type": "IMAGE"
7570 },
7571 {
7572 "id": 1134,
7573 "origin_id": -10,
7574 "origin_slot": 15,
7575 "target_id": 473,
7576 "target_slot": 2,
7577 "type": "BOOLEAN"
7578 },
7579 {
7580 "id": 1135,
7581 "origin_id": 473,
7582 "origin_slot": 0,
7583 "target_id": 470,
7584 "target_slot": 0,
7585 "type": "IMAGE"
7586 },
7587 {
7588 "id": 1136,
7589 "origin_id": 470,
7590 "origin_slot": 0,
7591 "target_id": 471,
7592 "target_slot": 0,
7593 "type": "INT"
7594 },
7595 {
7596 "id": 1137,
7597 "origin_id": 470,
7598 "origin_slot": 1,
7599 "target_id": 472,
7600 "target_slot": 0,
7601 "type": "INT"
7602 },
7603 {
7604 "id": 1138,
7605 "origin_id": 474,
7606 "origin_slot": 0,
7607 "target_id": 475,
7608 "target_slot": 1,
7609 "type": "SIGMAS"
7610 },
7611 {
7612 "id": 1139,
7613 "origin_id": 475,
7614 "origin_slot": 0,
7615 "target_id": 212,
7616 "target_slot": 3,
7617 "type": "SIGMAS"
7618 },
7619 {
7620 "id": 1140,
7621 "origin_id": -10,
7622 "origin_slot": 16,
7623 "target_id": 475,
7624 "target_slot": 2,
7625 "type": "BOOLEAN"
7626 }
7627 ],
7628 "extra": []
7629 },
7630 {
7631 "id": "5a45399a-b68e-5039-90e9-93ae815a69f1",
7632 "version": 1,
7633 "state": {
7634 "lastGroupId": 22,
7635 "lastNodeId": 470,
7636 "lastLinkId": 1132,
7637 "lastRerouteId": 0
7638 },
7639 "revision": 0,
7640 "config": [],
7641 "name": "Enlarge 2x with SeedVR2",
7642 "inputNode": {
7643 "id": -10,
7644 "bounding": [
7645 -260,
7646 120,
7647 170,
7648 160
7649 ]
7650 },
7651 "outputNode": {
7652 "id": -20,
7653 "bounding": [
7654 1100,
7655 120,
7656 170,
7657 80
7658 ]
7659 },
7660 "inputs": [
7661 {
7662 "id": "8dd17145-4fe8-5834-8151-9c977ef0ab6e",
7663 "name": "picture",
7664 "type": "IMAGE",
7665 "linkIds": [
7666 1109
7667 ],
7668 "localized_name": "picture",
7669 "pos": [
7670 0,
7671 0
7672 ]
7673 },
7674 {
7675 "id": "15db1f43-c6b7-5531-a354-54d92748709e",
7676 "name": "seed",
7677 "type": "INT",
7678 "linkIds": [
7679 1119
7680 ],
7681 "localized_name": "seed",
7682 "pos": [
7683 0,
7684 0
7685 ]
7686 },
7687 {
7688 "id": "8b28c901-3563-58fa-bc86-9e182f334dab",
7689 "name": "seedvr2 model",
7690 "type": "MODEL",
7691 "linkIds": [
7692 1113,
7693 1115
7694 ],
7695 "localized_name": "seedvr2 model",
7696 "pos": [
7697 0,
7698 0
7699 ]
7700 },
7701 {
7702 "id": "45092028-0346-50c4-837c-f77fee388519",
7703 "name": "seedvr2 vae",
7704 "type": "VAE",
7705 "linkIds": [
7706 1112,
7707 1121
7708 ],
7709 "localized_name": "seedvr2 vae",
7710 "pos": [
7711 0,
7712 0
7713 ]
7714 }
7715 ],
7716 "outputs": [
7717 {
7718 "id": "2c93f631-0cbe-5e5a-89dc-baba1651ee29",
7719 "name": "picture",
7720 "type": "IMAGE",
7721 "linkIds": [
7722 1124
7723 ],
7724 "localized_name": "picture",
7725 "pos": [
7726 0,
7727 0
7728 ]
7729 }
7730 ],
7731 "widgets": [],
7732 "nodes": [
7733 {
7734 "id": 117,
7735 "type": "AUSBOSS_NODES_Callout",
7736 "pos": [
7737 20,
7738 80
7739 ],
7740 "size": [
7741 460,
7742 150
7743 ],
7744 "flags": {
7745 "pinned": true
7746 },
7747 "order": 0,
7748 "mode": 0,
7749 "inputs": [
7750 {
7751 "localized_name": "text",
7752 "name": "text",
7753 "type": "STRING",
7754 "widget": {
7755 "name": "text"
7756 },
7757 "link": null
7758 }
7759 ],
7760 "outputs": [],
7761 "title": "Callout",
7762 "properties": {
7763 "Node name for S&R": "AUSBOSS_NODES_Callout",
7764 "cnr_id": "ausboss-nodes",
7765 "ver": "2.16.1"
7766 },
7767 "widgets_values": [
7768 "Your picture is made twice as big on each side.\n###### A 4 megapixel picture becomes 16 megapixels."
7769 ],
7770 "widgets_values_named": {
7771 "text": "Your picture is made twice as big on each side.\n###### A 4 megapixel picture becomes 16 megapixels."
7772 },
7773 "color": "#07090a",
7774 "bgcolor": "#07090a"
7775 },
7776 {
7777 "id": 110,
7778 "type": "AUSBOSS_NODES_ImageResize",
7779 "pos": [
7780 20,
7781 300
7782 ],
7783 "size": [
7784 460,
7785 330
7786 ],
7787 "flags": [],
7788 "order": 2,
7789 "mode": 0,
7790 "inputs": [
7791 {
7792 "localized_name": "image",
7793 "name": "image",
7794 "type": "IMAGE",
7795 "link": 1109
7796 },
7797 {
7798 "localized_name": "mask",
7799 "name": "mask",
7800 "shape": 7,
7801 "type": "MASK",
7802 "link": null
7803 },
7804 {
7805 "localized_name": "target_mode",
7806 "name": "target_mode",
7807 "type": "COMBO",
7808 "widget": {
7809 "name": "target_mode"
7810 },
7811 "link": null
7812 },
7813 {
7814 "localized_name": "width",
7815 "name": "width",
7816 "type": "INT",
7817 "link": null
7818 },
7819 {
7820 "localized_name": "height",
7821 "name": "height",
7822 "type": "INT",
7823 "link": null
7824 },
7825 {
7826 "localized_name": "edge_length",
7827 "name": "edge_length",
7828 "type": "INT",
7829 "widget": {
7830 "name": "edge_length"
7831 },
7832 "link": null
7833 },
7834 {
7835 "localized_name": "megapixels",
7836 "name": "megapixels",
7837 "type": "FLOAT",
7838 "widget": {
7839 "name": "megapixels"
7840 },
7841 "link": null
7842 },
7843 {
7844 "localized_name": "scale_factor",
7845 "name": "scale_factor",
7846 "type": "FLOAT",
7847 "widget": {
7848 "name": "scale_factor"
7849 },
7850 "link": null
7851 },
7852 {
7853 "localized_name": "keep_proportion",
7854 "name": "keep_proportion",
7855 "type": "COMBO",
7856 "widget": {
7857 "name": "keep_proportion"
7858 },
7859 "link": null
7860 },
7861 {
7862 "localized_name": "fill_color",
7863 "name": "fill_color",
7864 "type": "STRING",
7865 "widget": {
7866 "name": "fill_color"
7867 },
7868 "link": null
7869 },
7870 {
7871 "localized_name": "divisible_by",
7872 "name": "divisible_by",
7873 "type": "INT",
7874 "widget": {
7875 "name": "divisible_by"
7876 },
7877 "link": null
7878 },
7879 {
7880 "localized_name": "interpolation",
7881 "name": "interpolation",
7882 "type": "COMBO",
7883 "widget": {
7884 "name": "interpolation"
7885 },
7886 "link": null
7887 }
7888 ],
7889 "outputs": [
7890 {
7891 "localized_name": "image",
7892 "name": "image",
7893 "type": "IMAGE",
7894 "links": [
7895 1110,
7896 1123
7897 ]
7898 },
7899 {
7900 "localized_name": "mask",
7901 "name": "mask",
7902 "type": "MASK",
7903 "links": []
7904 },
7905 {
7906 "localized_name": "width",
7907 "name": "width",
7908 "type": "INT",
7909 "links": []
7910 },
7911 {
7912 "localized_name": "height",
7913 "name": "height",
7914 "type": "INT",
7915 "links": []
7916 }
7917 ],
7918 "title": "Image Resize - Twice the size on each side",
7919 "properties": {
7920 "Node name for S&R": "AUSBOSS_NODES_ImageResize",
7921 "cnr_id": "ausboss-nodes",
7922 "ver": "2.16.1"
7923 },
7924 "widgets_values": [
7925 "scale_factor",
7926 0,
7927 0,
7928 1024,
7929 1,
7930 2,
7931 "fit",
7932 "#000000",
7933 16,
7934 "lanczos"
7935 ],
7936 "widgets_values_named": {
7937 "target_mode": "scale_factor",
7938 "width": 0,
7939 "height": 0,
7940 "edge_length": 1024,
7941 "megapixels": 1,
7942 "scale_factor": 2,
7943 "keep_proportion": "fit",
7944 "fill_color": "#000000",
7945 "divisible_by": 16,
7946 "interpolation": "lanczos"
7947 },
7948 "color": "#14424d",
7949 "bgcolor": "#161f21"
7950 },
7951 {
7952 "id": 118,
7953 "type": "AUSBOSS_NODES_Callout",
7954 "pos": [
7955 560,
7956 80
7957 ],
7958 "size": [
7959 460,
7960 150
7961 ],
7962 "flags": {
7963 "pinned": true
7964 },
7965 "order": 1,
7966 "mode": 0,
7967 "inputs": [
7968 {
7969 "localized_name": "text",
7970 "name": "text",
7971 "type": "STRING",
7972 "widget": {
7973 "name": "text"
7974 },
7975 "link": null
7976 }
7977 ],
7978 "outputs": [],
7979 "title": "Callout",
7980 "properties": {
7981 "Node name for S&R": "AUSBOSS_NODES_Callout",
7982 "cnr_id": "ausboss-nodes",
7983 "ver": "2.16.1"
7984 },
7985 "widgets_values": [
7986 "One step of SeedVR2 3B sharpens it.\n###### Then its colors are set back to MiniMax-H3-Image's picture. Its seed is the **Seed**."
7987 ],
7988 "widgets_values_named": {
7989 "text": "One step of SeedVR2 3B sharpens it.\n###### Then its colors are set back to MiniMax-H3-Image's picture. Its seed is the **Seed**."
7990 },
7991 "color": "#07090a",
7992 "bgcolor": "#07090a"
7993 },
7994 {
7995 "id": 111,
7996 "type": "SeedVR2Preprocess",
7997 "pos": [
7998 560,
7999 300
8000 ],
8001 "size": [
8002 460,
8003 46
8004 ],
8005 "flags": [],
8006 "order": 3,
8007 "mode": 0,
8008 "inputs": [
8009 {
8010 "localized_name": "resized_images",
8011 "name": "resized_images",
8012 "type": "IMAGE",
8013 "link": 1110
8014 }
8015 ],
8016 "outputs": [
8017 {
8018 "localized_name": "images",
8019 "name": "images",
8020 "type": "IMAGE",
8021 "links": [
8022 1111
8023 ]
8024 }
8025 ],
8026 "title": "SeedVR2 Preprocess",
8027 "properties": {
8028 "Node name for S&R": "SeedVR2Preprocess",
8029 "cnr_id": "comfy-core",
8030 "ver": "0.41.0"
8031 },
8032 "color": "#14424d",
8033 "bgcolor": "#161f21"
8034 },
8035 {
8036 "id": 112,
8037 "type": "VAEEncodeTiled",
8038 "pos": [
8039 560,
8040 416
8041 ],
8042 "size": [
8043 460,
8044 154
8045 ],
8046 "flags": [],
8047 "order": 4,
8048 "mode": 0,
8049 "inputs": [
8050 {
8051 "localized_name": "pixels",
8052 "name": "pixels",
8053 "type": "IMAGE",
8054 "link": 1111
8055 },
8056 {
8057 "localized_name": "vae",
8058 "name": "vae",
8059 "type": "VAE",
8060 "link": 1112
8061 },
8062 {
8063 "localized_name": "tile_size",
8064 "name": "tile_size",
8065 "type": "INT",
8066 "widget": {
8067 "name": "tile_size"
8068 },
8069 "link": null
8070 },
8071 {
8072 "localized_name": "overlap",
8073 "name": "overlap",
8074 "type": "INT",
8075 "widget": {
8076 "name": "overlap"
8077 },
8078 "link": null
8079 },
8080 {
8081 "localized_name": "temporal_size",
8082 "name": "temporal_size",
8083 "type": "INT",
8084 "widget": {
8085 "name": "temporal_size"
8086 },
8087 "link": null
8088 },
8089 {
8090 "localized_name": "temporal_overlap",
8091 "name": "temporal_overlap",
8092 "type": "INT",
8093 "widget": {
8094 "name": "temporal_overlap"
8095 },
8096 "link": null
8097 }
8098 ],
8099 "outputs": [
8100 {
8101 "localized_name": "LATENT",
8102 "name": "LATENT",
8103 "type": "LATENT",
8104 "links": [
8105 1114,
8106 1118
8107 ]
8108 }
8109 ],
8110 "properties": {
8111 "Node name for S&R": "VAEEncodeTiled",
8112 "cnr_id": "comfy-core",
8113 "ver": "0.41.0"
8114 },
8115 "widgets_values": [
8116 512,
8117 128,
8118 4096,
8119 8
8120 ],
8121 "widgets_values_named": {
8122 "tile_size": 512,
8123 "overlap": 128,
8124 "temporal_size": 4096,
8125 "temporal_overlap": 8
8126 },
8127 "color": "#14424d",
8128 "bgcolor": "#161f21"
8129 },
8130 {
8131 "id": 113,
8132 "type": "SeedVR2Conditioning",
8133 "pos": [
8134 560,
8135 640
8136 ],
8137 "size": [
8138 460,
8139 66
8140 ],
8141 "flags": [],
8142 "order": 5,
8143 "mode": 0,
8144 "inputs": [
8145 {
8146 "localized_name": "model",
8147 "name": "model",
8148 "type": "MODEL",
8149 "link": 1113
8150 },
8151 {
8152 "localized_name": "latent",
8153 "name": "vae_conditioning",
8154 "type": "LATENT",
8155 "link": 1114
8156 }
8157 ],
8158 "outputs": [
8159 {
8160 "localized_name": "positive",
8161 "name": "positive",
8162 "type": "CONDITIONING",
8163 "links": [
8164 1116
8165 ]
8166 },
8167 {
8168 "localized_name": "negative",
8169 "name": "negative",
8170 "type": "CONDITIONING",
8171 "links": [
8172 1117
8173 ]
8174 }
8175 ],
8176 "title": "SeedVR2 Conditioning",
8177 "properties": {
8178 "Node name for S&R": "SeedVR2Conditioning",
8179 "cnr_id": "comfy-core",
8180 "ver": "0.41.0"
8181 },
8182 "color": "#14424d",
8183 "bgcolor": "#161f21"
8184 },
8185 {
8186 "id": 114,
8187 "type": "KSampler",
8188 "pos": [
8189 560,
8190 776
8191 ],
8192 "size": [
8193 460,
8194 262
8195 ],
8196 "flags": [],
8197 "order": 6,
8198 "mode": 0,
8199 "inputs": [
8200 {
8201 "localized_name": "model",
8202 "name": "model",
8203 "type": "MODEL",
8204 "link": 1115
8205 },
8206 {
8207 "localized_name": "positive",
8208 "name": "positive",
8209 "type": "CONDITIONING",
8210 "link": 1116
8211 },
8212 {
8213 "localized_name": "negative",
8214 "name": "negative",
8215 "type": "CONDITIONING",
8216 "link": 1117
8217 },
8218 {
8219 "localized_name": "latent_image",
8220 "name": "latent_image",
8221 "type": "LATENT",
8222 "link": 1118
8223 },
8224 {
8225 "localized_name": "seed",
8226 "name": "seed",
8227 "type": "INT",
8228 "widget": {
8229 "name": "seed"
8230 },
8231 "link": 1119
8232 },
8233 {
8234 "localized_name": "steps",
8235 "name": "steps",
8236 "type": "INT",
8237 "widget": {
8238 "name": "steps"
8239 },
8240 "link": null
8241 },
8242 {
8243 "localized_name": "cfg",
8244 "name": "cfg",
8245 "type": "FLOAT",
8246 "widget": {
8247 "name": "cfg"
8248 },
8249 "link": null
8250 },
8251 {
8252 "localized_name": "sampler_name",
8253 "name": "sampler_name",
8254 "type": "COMBO",
8255 "widget": {
8256 "name": "sampler_name"
8257 },
8258 "link": null
8259 },
8260 {
8261 "localized_name": "scheduler",
8262 "name": "scheduler",
8263 "type": "COMBO",
8264 "widget": {
8265 "name": "scheduler"
8266 },
8267 "link": null
8268 },
8269 {
8270 "localized_name": "denoise",
8271 "name": "denoise",
8272 "type": "FLOAT",
8273 "widget": {
8274 "name": "denoise"
8275 },
8276 "link": null
8277 }
8278 ],
8279 "outputs": [
8280 {
8281 "localized_name": "LATENT",
8282 "name": "LATENT",
8283 "type": "LATENT",
8284 "links": [
8285 1120
8286 ]
8287 }
8288 ],
8289 "title": "KSampler - 1 step, CFG 1",
8290 "properties": {
8291 "Node name for S&R": "KSampler",
8292 "cnr_id": "comfy-core",
8293 "ver": "0.41.0"
8294 },
8295 "widgets_values": [
8296 20261017,
8297 "fixed",
8298 1,
8299 1,
8300 "euler",
8301 "simple",
8302 1
8303 ],
8304 "widgets_values_named": {
8305 "seed": 20261017,
8306 "control_after_generate": "fixed",
8307 "steps": 1,
8308 "cfg": 1,
8309 "sampler_name": "euler",
8310 "scheduler": "simple",
8311 "denoise": 1
8312 },
8313 "color": "#14424d",
8314 "bgcolor": "#161f21"
8315 },
8316 {
8317 "id": 115,
8318 "type": "VAEDecodeTiled",
8319 "pos": [
8320 560,
8321 1108
8322 ],
8323 "size": [
8324 460,
8325 154
8326 ],
8327 "flags": [],
8328 "order": 7,
8329 "mode": 0,
8330 "inputs": [
8331 {
8332 "localized_name": "samples",
8333 "name": "samples",
8334 "type": "LATENT",
8335 "link": 1120
8336 },
8337 {
8338 "localized_name": "vae",
8339 "name": "vae",
8340 "type": "VAE",
8341 "link": 1121
8342 },
8343 {
8344 "localized_name": "tile_size",
8345 "name": "tile_size",
8346 "type": "INT",
8347 "widget": {
8348 "name": "tile_size"
8349 },
8350 "link": null
8351 },
8352 {
8353 "localized_name": "overlap",
8354 "name": "overlap",
8355 "type": "INT",
8356 "widget": {
8357 "name": "overlap"
8358 },
8359 "link": null
8360 },
8361 {
8362 "localized_name": "temporal_size",
8363 "name": "temporal_size",
8364 "type": "INT",
8365 "widget": {
8366 "name": "temporal_size"
8367 },
8368 "link": null
8369 },
8370 {
8371 "localized_name": "temporal_overlap",
8372 "name": "temporal_overlap",
8373 "type": "INT",
8374 "widget": {
8375 "name": "temporal_overlap"
8376 },
8377 "link": null
8378 }
8379 ],
8380 "outputs": [
8381 {
8382 "localized_name": "IMAGE",
8383 "name": "IMAGE",
8384 "type": "IMAGE",
8385 "links": [
8386 1122
8387 ]
8388 }
8389 ],
8390 "properties": {
8391 "Node name for S&R": "VAEDecodeTiled",
8392 "cnr_id": "comfy-core",
8393 "ver": "0.41.0"
8394 },
8395 "widgets_values": [
8396 512,
8397 128,
8398 4096,
8399 8
8400 ],
8401 "widgets_values_named": {
8402 "tile_size": 512,
8403 "overlap": 128,
8404 "temporal_size": 4096,
8405 "temporal_overlap": 8
8406 },
8407 "color": "#14424d",
8408 "bgcolor": "#161f21"
8409 },
8410 {
8411 "id": 116,
8412 "type": "SeedVR2PostProcessing",
8413 "pos": [
8414 560,
8415 1332
8416 ],
8417 "size": [
8418 460,
8419 82
8420 ],
8421 "flags": [],
8422 "order": 8,
8423 "mode": 0,
8424 "inputs": [
8425 {
8426 "localized_name": "images",
8427 "name": "images",
8428 "type": "IMAGE",
8429 "link": 1122
8430 },
8431 {
8432 "localized_name": "original_resized_images",
8433 "name": "original_resized_images",
8434 "type": "IMAGE",
8435 "link": 1123
8436 },
8437 {
8438 "localized_name": "color_correction_method",
8439 "name": "color_correction_method",
8440 "type": "COMBO",
8441 "widget": {
8442 "name": "color_correction_method"
8443 },
8444 "link": null
8445 }
8446 ],
8447 "outputs": [
8448 {
8449 "localized_name": "images",
8450 "name": "images",
8451 "type": "IMAGE",
8452 "links": [
8453 1124
8454 ]
8455 }
8456 ],
8457 "title": "SeedVR2 Post Processing - Colors back to MiniMax-H3-Image's",
8458 "properties": {
8459 "Node name for S&R": "SeedVR2PostProcessing",
8460 "cnr_id": "comfy-core",
8461 "ver": "0.41.0"
8462 },
8463 "widgets_values": [
8464 "lab"
8465 ],
8466 "widgets_values_named": {
8467 "color_correction_method": "lab"
8468 },
8469 "color": "#14424d",
8470 "bgcolor": "#161f21"
8471 }
8472 ],
8473 "groups": [
8474 {
8475 "id": 15,
8476 "title": "1 · Twice the size",
8477 "bounding": [
8478 0,
8479 0,
8480 500,
8481 650
8482 ],
8483 "color": "#1d5560",
8484 "flags": {
8485 "pinned": true
8486 }
8487 },
8488 {
8489 "id": 16,
8490 "title": "2 · SeedVR2 sharpens it",
8491 "bounding": [
8492 540,
8493 0,
8494 500,
8495 1434
8496 ],
8497 "color": "#1d5560",
8498 "flags": {
8499 "pinned": true
8500 }
8501 }
8502 ],
8503 "links": [
8504 {
8505 "id": 1109,
8506 "origin_id": -10,
8507 "origin_slot": 0,
8508 "target_id": 110,
8509 "target_slot": 0,
8510 "type": "IMAGE"
8511 },
8512 {
8513 "id": 1110,
8514 "origin_id": 110,
8515 "origin_slot": 0,
8516 "target_id": 111,
8517 "target_slot": 0,
8518 "type": "IMAGE"
8519 },
8520 {
8521 "id": 1111,
8522 "origin_id": 111,
8523 "origin_slot": 0,
8524 "target_id": 112,
8525 "target_slot": 0,
8526 "type": "IMAGE"
8527 },
8528 {
8529 "id": 1112,
8530 "origin_id": -10,
8531 "origin_slot": 3,
8532 "target_id": 112,
8533 "target_slot": 1,
8534 "type": "VAE"
8535 },
8536 {
8537 "id": 1113,
8538 "origin_id": -10,
8539 "origin_slot": 2,
8540 "target_id": 113,
8541 "target_slot": 0,
8542 "type": "MODEL"
8543 },
8544 {
8545 "id": 1114,
8546 "origin_id": 112,
8547 "origin_slot": 0,
8548 "target_id": 113,
8549 "target_slot": 1,
8550 "type": "LATENT"
8551 },
8552 {
8553 "id": 1115,
8554 "origin_id": -10,
8555 "origin_slot": 2,
8556 "target_id": 114,
8557 "target_slot": 0,
8558 "type": "MODEL"
8559 },
8560 {
8561 "id": 1116,
8562 "origin_id": 113,
8563 "origin_slot": 0,
8564 "target_id": 114,
8565 "target_slot": 1,
8566 "type": "CONDITIONING"
8567 },
8568 {
8569 "id": 1117,
8570 "origin_id": 113,
8571 "origin_slot": 1,
8572 "target_id": 114,
8573 "target_slot": 2,
8574 "type": "CONDITIONING"
8575 },
8576 {
8577 "id": 1118,
8578 "origin_id": 112,
8579 "origin_slot": 0,
8580 "target_id": 114,
8581 "target_slot": 3,
8582 "type": "LATENT"
8583 },
8584 {
8585 "id": 1119,
8586 "origin_id": -10,
8587 "origin_slot": 1,
8588 "target_id": 114,
8589 "target_slot": 4,
8590 "type": "INT"
8591 },
8592 {
8593 "id": 1120,
8594 "origin_id": 114,
8595 "origin_slot": 0,
8596 "target_id": 115,
8597 "target_slot": 0,
8598 "type": "LATENT"
8599 },
8600 {
8601 "id": 1121,
8602 "origin_id": -10,
8603 "origin_slot": 3,
8604 "target_id": 115,
8605 "target_slot": 1,
8606 "type": "VAE"
8607 },
8608 {
8609 "id": 1122,
8610 "origin_id": 115,
8611 "origin_slot": 0,
8612 "target_id": 116,
8613 "target_slot": 0,
8614 "type": "IMAGE"
8615 },
8616 {
8617 "id": 1123,
8618 "origin_id": 110,
8619 "origin_slot": 0,
8620 "target_id": 116,
8621 "target_slot": 1,
8622 "type": "IMAGE"
8623 },
8624 {
8625 "id": 1124,
8626 "origin_id": 116,
8627 "origin_slot": 0,
8628 "target_id": -20,
8629 "target_slot": 0,
8630 "type": "IMAGE"
8631 }
8632 ],
8633 "extra": []
8634 }
8635 ]
8636 },
8637 "config": [],
8638 "extra": {
8639 "ds": {
8640 "scale": 0.6000000000000001,
8641 "offset": [
8642 -458.2266982269287,
8643 -1206.1240202128097
8644 ]
8645 },
8646 "frontendVersion": "1.53.6"
8647 },
8648 "version": 0.4
8649}