工作流 /
图像工具 /
图像分割：WeVisOne-VL
图像分割：WeVisOne-VL
Image Segmentation: WeVisOne-VL
用自然语言指定要抠什么：给它一句如 human 的短语，WeVisOne-VL 就返回每个实例的 alpha 蒙版、框和点，预览里直接给出蒙版叠加、前景与背景抠图，蒙版可以直接接背景去除、遮罩或合成流程。
图像工具
9 个节点 1 个输入 ComfyUI ≥ 0.45.0 ComfyUI 31 KB
需要的权重
这个工作流引用了 1 个权重文件，其中 1 个能在本站模型库里找到。
wevisone_vl_2b_int8_convrot.safetensors
在模型库查看
节点构成
PreviewImage
× 3
MarkdownNote
× 2
LoadImage
× 1
1d74ebdb-3aaa-4d28-84fc-abf8b61daae2
× 1
MaskPreview
× 1
DrawBBoxes
× 1
官方标注的模型
WeVisOne-VL
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "41e71571-7413-428e-a913-4073adce9661",
3 "revision": 0,
4 "last_node_id": 49,
5 "last_link_id": 53,
6 "nodes": [
7 {
8 "id": 10,
9 "type": "LoadImage",
10 "pos": [
11 -40,
12 30
13 ],
14 "size": [
15 320,
16 340
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "IMAGE",
25 "type": "IMAGE",
26 "slot_index": 0,
27 "links": [
28 22,
29 48
30 ]
31 },
32 {
33 "name": "MASK",
34 "type": "MASK",
35 "slot_index": 1,
36 "links": []
37 }
38 ],
39 "properties": {
40 "Node name for S&R": "LoadImage"
41 },
42 "widgets_values": [
43 "utility_image_segment_wevisone_vl_input.png",
44 "image"
45 ],
46 "widgets_values_named": {
47 "image": "utility_image_segment_wevisone_vl_input.png",
48 "upload": "image"
49 }
50 },
51 {
52 "id": 36,
53 "type": "1d74ebdb-3aaa-4d28-84fc-abf8b61daae2",
54 "pos": [
55 340,
56 30
57 ],
58 "size": [
59 330,
60 380
61 ],
62 "flags": [],
63 "order": 3,
64 "mode": 0,
65 "inputs": [
66 {
67 "name": "image",
68 "type": "IMAGE",
69 "link": 22
70 },
71 {
72 "name": "bboxes",
73 "shape": 7,
74 "type": "BOUNDING_BOX",
75 "link": null
76 }
77 ],
78 "outputs": [
79 {
80 "name": "masks",
81 "type": "MASK",
82 "links": [
83 46
84 ]
85 },
86 {
87 "name": "bboxes",
88 "type": "BOUNDING_BOX",
89 "links": [
90 47
91 ]
92 },
93 {
94 "name": "points",
95 "type": "STRING",
96 "links": null
97 },
98 {
99 "name": "text",
100 "type": "STRING",
101 "links": []
102 },
103 {
104 "label": "foreground_image",
105 "name": "IMAGE",
106 "type": "IMAGE",
107 "links": [
108 51
109 ]
110 },
111 {
112 "label": "background_image",
113 "name": "IMAGE_1",
114 "type": "IMAGE",
115 "links": [
116 52
117 ]
118 }
119 ],
120 "properties": {
121 "previewExposures": []
122 },
123 "widgets_values": [
124 "segmentation",
125 "human",
126 false,
127 4096,
128 "wevisone_vl_2b_int8_convrot.safetensors"
129 ],
130 "widgets_values_named": {
131 "task": "segmentation",
132 "query": "human",
133 "individual_masks": false,
134 "max_length": 4096,
135 "clip_name": "wevisone_vl_2b_int8_convrot.safetensors"
136 }
137 },
138 {
139 "id": 41,
140 "type": "MaskPreview",
141 "pos": [
142 880,
143 -320
144 ],
145 "size": [
146 310,
147 290
148 ],
149 "flags": [],
150 "order": 4,
151 "mode": 0,
152 "inputs": [
153 {
154 "name": "mask",
155 "type": "MASK",
156 "link": 46
157 }
158 ],
159 "outputs": [
160 {
161 "name": "mask",
162 "type": "MASK",
163 "links": null
164 }
165 ],
166 "properties": {
167 "Node name for S&R": "MaskPreview"
168 }
169 },
170 {
171 "id": 42,
172 "type": "DrawBBoxes",
173 "pos": [
174 880,
175 20
176 ],
177 "size": [
178 150,
179 50
180 ],
181 "flags": [],
182 "order": 5,
183 "mode": 0,
184 "inputs": [
185 {
186 "name": "bboxes",
187 "type": "BOUNDING_BOX",
188 "link": 47
189 },
190 {
191 "name": "image",
192 "shape": 7,
193 "type": "IMAGE",
194 "link": 48
195 }
196 ],
197 "outputs": [
198 {
199 "name": "out_image",
200 "type": "IMAGE",
201 "links": [
202 50
203 ]
204 }
205 ],
206 "properties": {
207 "Node name for S&R": "DrawBBoxes"
208 }
209 },
210 {
211 "id": 44,
212 "type": "PreviewImage",
213 "pos": [
214 1210,
215 20
216 ],
217 "size": [
218 360,
219 410
220 ],
221 "flags": [],
222 "order": 8,
223 "mode": 0,
224 "inputs": [
225 {
226 "name": "images",
227 "type": "IMAGE",
228 "link": 50
229 }
230 ],
231 "outputs": [
232 {
233 "name": "images",
234 "type": "IMAGE",
235 "links": null
236 }
237 ],
238 "properties": {
239 "Node name for S&R": "PreviewImage"
240 }
241 },
242 {
243 "id": 45,
244 "type": "PreviewImage",
245 "pos": [
246 880,
247 140
248 ],
249 "size": [
250 270,
251 290
252 ],
253 "flags": [],
254 "order": 6,
255 "mode": 0,
256 "inputs": [
257 {
258 "name": "images",
259 "type": "IMAGE",
260 "link": 51
261 }
262 ],
263 "outputs": [
264 {
265 "name": "images",
266 "type": "IMAGE",
267 "links": []
268 }
269 ],
270 "properties": {
271 "Node name for S&R": "PreviewImage"
272 }
273 },
274 {
275 "id": 46,
276 "type": "PreviewImage",
277 "pos": [
278 880,
279 480
280 ],
281 "size": [
282 270,
283 390
284 ],
285 "flags": [],
286 "order": 7,
287 "mode": 0,
288 "inputs": [
289 {
290 "name": "images",
291 "type": "IMAGE",
292 "link": 52
293 }
294 ],
295 "outputs": [
296 {
297 "name": "images",
298 "type": "IMAGE",
299 "links": []
300 }
301 ],
302 "properties": {
303 "Node name for S&R": "PreviewImage"
304 }
305 },
306 {
307 "id": 48,
308 "type": "MarkdownNote",
309 "pos": [
310 -490,
311 30
312 ],
313 "size": [
314 410,
315 890
316 ],
317 "flags": [],
318 "order": 1,
319 "mode": 0,
320 "inputs": [],
321 "outputs": [],
322 "properties": [],
323 "widgets_values": [
324 "Guide: [Subgraph](https://docs.comfy.org/interface/features/subgraph)\n\n\n## How to use\n\n**Load Image** → **WeVisOne-VL Segment** (subgraph) → previews. One pass returns alpha mattes, boxes, points and the raw answer; the graph previews the matte, the boxes drawn over the source, and the foreground and background cutouts on transparency, so the matte drops straight into a background-removal, masking or compositing chain.\n\n- `query` — the graph ships `segmentation` with `human`. Put whatever you want cut out in there as a short noun phrase; a full sentence makes `segmentation` ramble.\n- `task` — ten tasks, listed with what each returns in the model note next to this one.\n- `individual_masks` — one matte per instance instead of the union of them.\n- The `MASK` output wires straight into mask nodes (`Invert Mask`, `Grow Mask`, `Join Image with Alpha`, …), so the matte can drive any matting or compositing branch.\n\n## About this model\n\nWeVisOne-VL is a compact vision-language model that answers in shapes instead of prose: it locates what a text query names and returns boxes, points or per-instance alpha mattes in one pass. Native in ComfyUI since #17632.\n\n## Tasks\n\n`task` rewrites your query into the instruction the checkpoint was trained on:\n\n- `segmentation` — alpha matte per instance plus its box (this graph).\n- `detection` — boxes only, one label per line in the query.\n- `referring` — boxes for a describing phrase (\"the cup on the left\"), not just a category.\n- `point` — centre points as JSON instead of boxes.\n- `count_box` — counts the instances and boxes each one.\n- `count_point` — counts them and marks each with a point.\n- `ocr` — boxes every word in the picture.\n- `ocr_grounding` — finds the one string you put in the query and boxes it.\n- `visual_prompt` — takes reference boxes on the optional `bboxes` input and finds similar objects; wire a Bounding Box source in first.\n- `freeform` — sends the query as the whole prompt, for anything else this model can answer.\n\n`individual_masks` splits the mattes per instance instead of one union per image; `max_length` caps the answer length."
325 ],
326 "widgets_values_named": {
327 "text": "Guide: [Subgraph](https://docs.comfy.org/interface/features/subgraph)\n\n\n## How to use\n\n**Load Image** → **WeVisOne-VL Segment** (subgraph) → previews. One pass returns alpha mattes, boxes, points and the raw answer; the graph previews the matte, the boxes drawn over the source, and the foreground and background cutouts on transparency, so the matte drops straight into a background-removal, masking or compositing chain.\n\n- `query` — the graph ships `segmentation` with `human`. Put whatever you want cut out in there as a short noun phrase; a full sentence makes `segmentation` ramble.\n- `task` — ten tasks, listed with what each returns in the model note next to this one.\n- `individual_masks` — one matte per instance instead of the union of them.\n- The `MASK` output wires straight into mask nodes (`Invert Mask`, `Grow Mask`, `Join Image with Alpha`, …), so the matte can drive any matting or compositing branch.\n\n## About this model\n\nWeVisOne-VL is a compact vision-language model that answers in shapes instead of prose: it locates what a text query names and returns boxes, points or per-instance alpha mattes in one pass. Native in ComfyUI since #17632.\n\n## Tasks\n\n`task` rewrites your query into the instruction the checkpoint was trained on:\n\n- `segmentation` — alpha matte per instance plus its box (this graph).\n- `detection` — boxes only, one label per line in the query.\n- `referring` — boxes for a describing phrase (\"the cup on the left\"), not just a category.\n- `point` — centre points as JSON instead of boxes.\n- `count_box` — counts the instances and boxes each one.\n- `count_point` — counts them and marks each with a point.\n- `ocr` — boxes every word in the picture.\n- `ocr_grounding` — finds the one string you put in the query and boxes it.\n- `visual_prompt` — takes reference boxes on the optional `bboxes` input and finds similar objects; wire a Bounding Box source in first.\n- `freeform` — sends the query as the whole prompt, for anything else this model can answer.\n\n`individual_masks` splits the mattes per instance instead of one union per image; `max_length` caps the answer length."
328 },
329 "color": "#222",
330 "bgcolor": "#000"
331 },
332 {
333 "id": 49,
334 "type": "MarkdownNote",
335 "pos": [
336 -900,
337 30
338 ],
339 "size": [
340 390,
341 880
342 ],
343 "flags": [],
344 "order": 2,
345 "mode": 0,
346 "inputs": [],
347 "outputs": [],
348 "title": "Note: Models & tasks",
349 "properties": [],
350 "widgets_values": [
351 "\n## Input Assets\n\n- [utility_image_segment_wevisone_vl_input.png](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/utility_image_segment_wevisone_vl_input.png)\n\n## Models\n\n- [Hugging Face:Comfy-Org/WeVisOne-VL](https://huggingface.co/Comfy-Org/WeVisOne-VL)\n\n**ModelScope mirrors** (alternative repo location)\n\n- [ModelScope:Comfy-Org/WeVisOne-VL](https://modelscope.cn/models/Comfy-Org/WeVisOne-VL)\n\n**text_encoders**\n\n- [wevisone_vl_2b_int8_convrot.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_2b_int8_convrot.safetensors) (2.69 GB) — loaded by this graph\n- [wevisone_vl_2b_bf16.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_2b_bf16.safetensors) (4.58 GB)\n- [wevisone_vl_4b_int8_convrot.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_4b_int8_convrot.safetensors) (4.93 GB)\n- [wevisone_vl_4b_bf16.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_4b_bf16.safetensors) (9.03 GB)\n\nSwap weights with `clip_name` in `CLIPLoader` (`stable_diffusion` type): the 4B pair is heavier and slower, the int8 files fit smaller GPUs.\n\n## Model Storage Location\n\n```\n📂 ComfyUI/\n├── 📂 models/\n│ └── 📂 text_encoders/\n│ └── wevisone_vl_2b_int8_convrot.safetensors\n```\n\n## Report Issue\n\nNote: Please update ComfyUI first ([guide](https://docs.comfy.org/installation/update_comfyui)) and prepare required models. Desktop/Cloud updates follow stable releases, so some nightly-supported models may not be available yet.\n\n- Cannot run / runtime errors: [ComfyUI/issues](https://github.com/comfyanonymous/ComfyUI/issues)\n- UI / frontend issues: [ComfyUI_frontend/issues](https://github.com/Comfy-Org/ComfyUI_frontend/issues)\n- Workflow issues: [workflow_templates/issues](https://github.com/Comfy-Org/workflow_templates/issues)\n\n"
352 ],
353 "widgets_values_named": {
354 "text": "\n## Input Assets\n\n- [utility_image_segment_wevisone_vl_input.png](https://raw.githubusercontent.com/Comfy-Org/workflow_templates/refs/heads/main/input/utility_image_segment_wevisone_vl_input.png)\n\n## Models\n\n- [Hugging Face:Comfy-Org/WeVisOne-VL](https://huggingface.co/Comfy-Org/WeVisOne-VL)\n\n**ModelScope mirrors** (alternative repo location)\n\n- [ModelScope:Comfy-Org/WeVisOne-VL](https://modelscope.cn/models/Comfy-Org/WeVisOne-VL)\n\n**text_encoders**\n\n- [wevisone_vl_2b_int8_convrot.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_2b_int8_convrot.safetensors) (2.69 GB) — loaded by this graph\n- [wevisone_vl_2b_bf16.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_2b_bf16.safetensors) (4.58 GB)\n- [wevisone_vl_4b_int8_convrot.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_4b_int8_convrot.safetensors) (4.93 GB)\n- [wevisone_vl_4b_bf16.safetensors](https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_4b_bf16.safetensors) (9.03 GB)\n\nSwap weights with `clip_name` in `CLIPLoader` (`stable_diffusion` type): the 4B pair is heavier and slower, the int8 files fit smaller GPUs.\n\n## Model Storage Location\n\n```\n📂 ComfyUI/\n├── 📂 models/\n│ └── 📂 text_encoders/\n│ └── wevisone_vl_2b_int8_convrot.safetensors\n```\n\n## Report Issue\n\nNote: Please update ComfyUI first ([guide](https://docs.comfy.org/installation/update_comfyui)) and prepare required models. Desktop/Cloud updates follow stable releases, so some nightly-supported models may not be available yet.\n\n- Cannot run / runtime errors: [ComfyUI/issues](https://github.com/comfyanonymous/ComfyUI/issues)\n- UI / frontend issues: [ComfyUI_frontend/issues](https://github.com/Comfy-Org/ComfyUI_frontend/issues)\n- Workflow issues: [workflow_templates/issues](https://github.com/Comfy-Org/workflow_templates/issues)\n\n"
355 },
356 "color": "#222",
357 "bgcolor": "#000"
358 }
359 ],
360 "links": [
361 [
362 22,
363 10,
364 0,
365 36,
366 0,
367 "IMAGE"
368 ],
369 [
370 46,
371 36,
372 0,
373 41,
374 0,
375 "MASK"
376 ],
377 [
378 47,
379 36,
380 1,
381 42,
382 0,
383 "BOUNDING_BOX"
384 ],
385 [
386 48,
387 10,
388 0,
389 42,
390 1,
391 "IMAGE"
392 ],
393 [
394 50,
395 42,
396 0,
397 44,
398 0,
399 "IMAGE"
400 ],
401 [
402 51,
403 36,
404 4,
405 45,
406 0,
407 "IMAGE"
408 ],
409 [
410 52,
411 36,
412 5,
413 46,
414 0,
415 "IMAGE"
416 ]
417 ],
418 "groups": [],
419 "definitions": {
420 "subgraphs": [
421 {
422 "id": "1d74ebdb-3aaa-4d28-84fc-abf8b61daae2",
423 "version": 1,
424 "state": {
425 "lastGroupId": 3,
426 "lastNodeId": 49,
427 "lastLinkId": 53,
428 "lastRerouteId": 0
429 },
430 "revision": 0,
431 "config": [],
432 "name": "WeVisOne-VL Segment",
433 "inputNode": {
434 "id": -10,
435 "bounding": [
436 440,
437 60,
438 144.369140625,
439 188
440 ]
441 },
442 "outputNode": {
443 "id": -20,
444 "bounding": [
445 1530,
446 -30,
447 156.169921875,
448 168
449 ]
450 },
451 "inputs": [
452 {
453 "id": "41da9b7f-48ba-4b70-a2bf-e3f69ede4e25",
454 "name": "image",
455 "type": "IMAGE",
456 "linkIds": [
457 2,
458 39,
459 43
460 ],
461 "localized_name": "image",
462 "pos": [
463 560.369140625,
464 84
465 ]
466 },
467 {
468 "id": "5f5ec8cb-1ef6-42b5-b30b-4372c13bc26c",
469 "name": "bboxes",
470 "type": "BOUNDING_BOX",
471 "linkIds": [
472 26
473 ],
474 "pos": [
475 560.369140625,
476 104
477 ]
478 },
479 {
480 "id": "afac0e56-0c26-4fe5-8fed-213a4cdd2276",
481 "name": "task",
482 "type": "COMBO",
483 "linkIds": [
484 27
485 ],
486 "pos": [
487 560.369140625,
488 124
489 ]
490 },
491 {
492 "id": "054feb94-47c6-4423-9ea3-f74641388cd3",
493 "name": "query",
494 "type": "STRING",
495 "linkIds": [
496 28
497 ],
498 "pos": [
499 560.369140625,
500 144
501 ]
502 },
503 {
504 "id": "49adf9b1-9086-4122-863e-9d9973edba30",
505 "name": "individual_masks",
506 "type": "BOOLEAN",
507 "linkIds": [
508 29
509 ],
510 "pos": [
511 560.369140625,
512 164
513 ]
514 },
515 {
516 "id": "9cc14703-45c9-46b3-b8c4-dafb8fee92bc",
517 "name": "max_length",
518 "type": "INT",
519 "linkIds": [
520 30
521 ],
522 "pos": [
523 560.369140625,
524 184
525 ]
526 },
527 {
528 "id": "f3a6e151-1934-4465-87ee-2fe0c258c75d",
529 "name": "clip_name",
530 "type": "COMBO",
531 "linkIds": [
532 31
533 ],
534 "pos": [
535 560.369140625,
536 204
537 ]
538 }
539 ],
540 "outputs": [
541 {
542 "id": "0b4ff730-f70a-4f05-a729-1c15054857c0",
543 "name": "masks",
544 "type": "MASK",
545 "linkIds": [
546 4
547 ],
548 "localized_name": "masks",
549 "pos": [
550 1554,
551 -6
552 ]
553 },
554 {
555 "id": "335cb944-e410-4821-aed1-265083396471",
556 "name": "bboxes",
557 "type": "BOUNDING_BOX",
558 "linkIds": [
559 32
560 ],
561 "pos": [
562 1554,
563 14
564 ]
565 },
566 {
567 "id": "7f527adf-de95-40e6-a195-cf2f3aae2677",
568 "name": "points",
569 "type": "STRING",
570 "linkIds": [
571 33
572 ],
573 "pos": [
574 1554,
575 34
576 ]
577 },
578 {
579 "id": "1968209f-bb5a-4dae-912e-694eea2b8e85",
580 "name": "text",
581 "type": "STRING",
582 "linkIds": [
583 34
584 ],
585 "pos": [
586 1554,
587 54
588 ]
589 },
590 {
591 "id": "21db7756-19d8-4577-a75b-a1d6d25d7d66",
592 "name": "IMAGE",
593 "type": "IMAGE",
594 "linkIds": [
595 38
596 ],
597 "label": "foreground_image",
598 "pos": [
599 1554,
600 74
601 ]
602 },
603 {
604 "id": "5a722e81-f599-4c83-b966-4be78f583c06",
605 "name": "IMAGE_1",
606 "type": "IMAGE",
607 "linkIds": [
608 44
609 ],
610 "label": "background_image",
611 "pos": [
612 1554,
613 94
614 ]
615 }
616 ],
617 "widgets": [],
618 "nodes": [
619 {
620 "id": 11,
621 "type": "WeVisOneDetect",
622 "pos": [
623 740,
624 -30
625 ],
626 "size": [
627 430,
628 350
629 ],
630 "flags": [],
631 "order": 1,
632 "mode": 0,
633 "showAdvanced": true,
634 "inputs": [
635 {
636 "localized_name": "clip",
637 "name": "clip",
638 "type": "CLIP",
639 "link": 1
640 },
641 {
642 "localized_name": "image",
643 "name": "image",
644 "type": "IMAGE",
645 "link": 2
646 },
647 {
648 "localized_name": "bboxes",
649 "name": "bboxes",
650 "shape": 7,
651 "type": "BOUNDING_BOX",
652 "link": 26
653 },
654 {
655 "localized_name": "task",
656 "name": "task",
657 "type": "COMBO",
658 "widget": {
659 "name": "task"
660 },
661 "link": 27
662 },
663 {
664 "localized_name": "query",
665 "name": "query",
666 "type": "STRING",
667 "widget": {
668 "name": "query"
669 },
670 "link": 28
671 },
672 {
673 "localized_name": "individual_masks",
674 "name": "individual_masks",
675 "type": "BOOLEAN",
676 "widget": {
677 "name": "individual_masks"
678 },
679 "link": 29
680 },
681 {
682 "localized_name": "max_length",
683 "name": "max_length",
684 "type": "INT",
685 "widget": {
686 "name": "max_length"
687 },
688 "link": 30
689 }
690 ],
691 "outputs": [
692 {
693 "localized_name": "masks",
694 "name": "masks",
695 "type": "MASK",
696 "slot_index": 0,
697 "links": [
698 4,
699 37,
700 42
701 ]
702 },
703 {
704 "localized_name": "bboxes",
705 "name": "bboxes",
706 "type": "BOUNDING_BOX",
707 "slot_index": 1,
708 "links": [
709 32
710 ]
711 },
712 {
713 "localized_name": "points",
714 "name": "points",
715 "type": "STRING",
716 "slot_index": 2,
717 "links": [
718 33
719 ]
720 },
721 {
722 "localized_name": "text",
723 "name": "text",
724 "type": "STRING",
725 "slot_index": 3,
726 "links": [
727 34
728 ]
729 }
730 ],
731 "properties": {
732 "Node name for S&R": "WeVisOneDetect"
733 },
734 "widgets_values": [
735 "segmentation",
736 "human",
737 false,
738 4096
739 ],
740 "widgets_values_named": {
741 "task": "segmentation",
742 "query": "human",
743 "individual_masks": false,
744 "max_length": 4096
745 }
746 },
747 {
748 "id": 1,
749 "type": "CLIPLoader",
750 "pos": [
751 750,
752 390
753 ],
754 "size": [
755 420,
756 120
757 ],
758 "flags": [],
759 "order": 0,
760 "mode": 0,
761 "inputs": [
762 {
763 "localized_name": "clip_name",
764 "name": "clip_name",
765 "type": "COMBO",
766 "widget": {
767 "name": "clip_name"
768 },
769 "link": 31
770 }
771 ],
772 "outputs": [
773 {
774 "localized_name": "CLIP",
775 "name": "CLIP",
776 "type": "CLIP",
777 "slot_index": 0,
778 "links": [
779 1
780 ]
781 }
782 ],
783 "properties": {
784 "Node name for S&R": "CLIPLoader",
785 "models": [
786 {
787 "name": "wevisone_vl_2b_int8_convrot.safetensors",
788 "url": "https://huggingface.co/Comfy-Org/WeVisOne-VL/resolve/main/text_encoders/wevisone_vl_2b_int8_convrot.safetensors",
789 "directory": "text_encoders"
790 }
791 ]
792 },
793 "widgets_values": [
794 "wevisone_vl_2b_int8_convrot.safetensors",
795 "stable_diffusion",
796 "default"
797 ],
798 "widgets_values_named": {
799 "clip_name": "wevisone_vl_2b_int8_convrot.safetensors",
800 "type": "stable_diffusion",
801 "device": "default"
802 }
803 },
804 {
805 "id": 37,
806 "type": "InvertMask",
807 "pos": [
808 1220,
809 130
810 ],
811 "size": [
812 180,
813 30
814 ],
815 "flags": [],
816 "order": 2,
817 "mode": 0,
818 "inputs": [
819 {
820 "localized_name": "mask",
821 "name": "mask",
822 "type": "MASK",
823 "link": 37
824 }
825 ],
826 "outputs": [
827 {
828 "localized_name": "MASK",
829 "name": "MASK",
830 "type": "MASK",
831 "slot_index": 0,
832 "links": [
833 36
834 ]
835 }
836 ],
837 "properties": {
838 "Node name for S&R": "InvertMask"
839 }
840 },
841 {
842 "id": 38,
843 "type": "JoinImageWithAlpha",
844 "pos": [
845 1220,
846 210
847 ],
848 "size": [
849 200,
850 50
851 ],
852 "flags": [],
853 "order": 3,
854 "mode": 0,
855 "inputs": [
856 {
857 "localized_name": "image",
858 "name": "image",
859 "type": "IMAGE",
860 "link": 39
861 },
862 {
863 "localized_name": "alpha",
864 "name": "alpha",
865 "type": "MASK",
866 "link": 36
867 }
868 ],
869 "outputs": [
870 {
871 "localized_name": "IMAGE",
872 "name": "IMAGE",
873 "type": "IMAGE",
874 "slot_index": 0,
875 "links": [
876 38
877 ]
878 }
879 ],
880 "properties": {
881 "Node name for S&R": "JoinImageWithAlpha"
882 }
883 },
884 {
885 "id": 39,
886 "type": "JoinImageWithAlpha",
887 "pos": [
888 1220,
889 320
890 ],
891 "size": [
892 200,
893 50
894 ],
895 "flags": [],
896 "order": 4,
897 "mode": 0,
898 "inputs": [
899 {
900 "localized_name": "image",
901 "name": "image",
902 "type": "IMAGE",
903 "link": 43
904 },
905 {
906 "localized_name": "alpha",
907 "name": "alpha",
908 "type": "MASK",
909 "link": 42
910 }
911 ],
912 "outputs": [
913 {
914 "localized_name": "IMAGE",
915 "name": "IMAGE",
916 "type": "IMAGE",
917 "slot_index": 0,
918 "links": [
919 44
920 ]
921 }
922 ],
923 "properties": {
924 "Node name for S&R": "JoinImageWithAlpha"
925 }
926 }
927 ],
928 "groups": [],
929 "links": [
930 {
931 "id": 1,
932 "origin_id": 1,
933 "origin_slot": 0,
934 "target_id": 11,
935 "target_slot": 0,
936 "type": "CLIP"
937 },
938 {
939 "id": 2,
940 "origin_id": -10,
941 "origin_slot": 0,
942 "target_id": 11,
943 "target_slot": 1,
944 "type": "IMAGE"
945 },
946 {
947 "id": 4,
948 "origin_id": 11,
949 "origin_slot": 0,
950 "target_id": -20,
951 "target_slot": 0,
952 "type": "MASK"
953 },
954 {
955 "id": 26,
956 "origin_id": -10,
957 "origin_slot": 1,
958 "target_id": 11,
959 "target_slot": 2,
960 "type": "BOUNDING_BOX"
961 },
962 {
963 "id": 27,
964 "origin_id": -10,
965 "origin_slot": 2,
966 "target_id": 11,
967 "target_slot": 3,
968 "type": "COMBO"
969 },
970 {
971 "id": 28,
972 "origin_id": -10,
973 "origin_slot": 3,
974 "target_id": 11,
975 "target_slot": 4,
976 "type": "STRING"
977 },
978 {
979 "id": 29,
980 "origin_id": -10,
981 "origin_slot": 4,
982 "target_id": 11,
983 "target_slot": 5,
984 "type": "BOOLEAN"
985 },
986 {
987 "id": 30,
988 "origin_id": -10,
989 "origin_slot": 5,
990 "target_id": 11,
991 "target_slot": 6,
992 "type": "INT"
993 },
994 {
995 "id": 31,
996 "origin_id": -10,
997 "origin_slot": 6,
998 "target_id": 1,
999 "target_slot": 0,
1000 "type": "COMBO"
1001 },
1002 {
1003 "id": 32,
1004 "origin_id": 11,
1005 "origin_slot": 1,
1006 "target_id": -20,
1007 "target_slot": 1,
1008 "type": "BOUNDING_BOX"
1009 },
1010 {
1011 "id": 33,
1012 "origin_id": 11,
1013 "origin_slot": 2,
1014 "target_id": -20,
1015 "target_slot": 2,
1016 "type": "STRING"
1017 },
1018 {
1019 "id": 34,
1020 "origin_id": 11,
1021 "origin_slot": 3,
1022 "target_id": -20,
1023 "target_slot": 3,
1024 "type": "STRING"
1025 },
1026 {
1027 "id": 36,
1028 "origin_id": 37,
1029 "origin_slot": 0,
1030 "target_id": 38,
1031 "target_slot": 1,
1032 "type": "MASK"
1033 },
1034 {
1035 "id": 37,
1036 "origin_id": 11,
1037 "origin_slot": 0,
1038 "target_id": 37,
1039 "target_slot": 0,
1040 "type": "MASK"
1041 },
1042 {
1043 "id": 38,
1044 "origin_id": 38,
1045 "origin_slot": 0,
1046 "target_id": -20,
1047 "target_slot": 4,
1048 "type": "IMAGE"
1049 },
1050 {
1051 "id": 39,
1052 "origin_id": -10,
1053 "origin_slot": 0,
1054 "target_id": 38,
1055 "target_slot": 0,
1056 "type": "IMAGE"
1057 },
1058 {
1059 "id": 42,
1060 "origin_id": 11,
1061 "origin_slot": 0,
1062 "target_id": 39,
1063 "target_slot": 1,
1064 "type": "MASK"
1065 },
1066 {
1067 "id": 43,
1068 "origin_id": -10,
1069 "origin_slot": 0,
1070 "target_id": 39,
1071 "target_slot": 0,
1072 "type": "IMAGE"
1073 },
1074 {
1075 "id": 44,
1076 "origin_id": 39,
1077 "origin_slot": 0,
1078 "target_id": -20,
1079 "target_slot": 5,
1080 "type": "IMAGE"
1081 }
1082 ],
1083 "extra": []
1084 }
1085 ]
1086 },
1087 "config": [],
1088 "extra": {
1089 "frontendVersion": "1.59.1",
1090 "VHS_latentpreview": false,
1091 "VHS_latentpreviewrate": 0,
1092 "VHS_MetadataImage": true,
1093 "VHS_KeepIntermediate": true,
1094 "ds": {
1095 "scale": 1.1518037665705803,
1096 "offset": [
1097 987.482593496577,
1098 309.9845773593652
1099 ]
1100 }
1101 },
1102 "version": 0.4
1103}