工作流 /
LLM /
OpenAI：ChatGPT
OpenAI：ChatGPT
OpenAI: ChatGPT
与OpenAI的高级语言模型互动进行智能对话。
LLM
4 个节点 1 个输入 32 次使用 ComfyUI ≥ 0.28.1 ComfyUI 4 KB
节点构成
OpenAIChatConfig
× 1
OpenAIChatNode
× 1
LoadImage
× 1
SaveText
× 1
官方标注的模型
ChatGPT
工作流 JSON
和上面「下载工作流 JSON」是同一份文件。
1{
2 "id": "083c898c-0a6e-4f9a-b9c1-7e8ce36506da",
3 "revision": 0,
4 "last_node_id": 19,
5 "last_link_id": 15,
6 "nodes": [
7 {
8 "id": 4,
9 "type": "OpenAIChatConfig",
10 "pos": [
11 -6240,
12 2510
13 ],
14 "size": [
15 390,
16 230
17 ],
18 "flags": [],
19 "order": 0,
20 "mode": 0,
21 "inputs": [],
22 "outputs": [
23 {
24 "name": "OPENAI_CHAT_CONFIG",
25 "type": "OPENAI_CHAT_CONFIG",
26 "links": [
27 3
28 ]
29 }
30 ],
31 "properties": {
32 "Node name for S&R": "OpenAIChatConfig",
33 "cnr_id": "comfy-core",
34 "ver": "0.20.1"
35 },
36 "widgets_values": [
37 "auto",
38 4096,
39 "You are an AI specialized in image-to-text prompt generation. Analyze the given image carefully and describe it in a detailed, structured, and prompt-friendly format. Include elements such as subject, clothing, style, environment, lighting, composition, color palette, mood, camera angle, and any other relevant visual details. Your output should resemble a high-quality text-to-image prompt that can help recreate the image. Use concise, descriptive language. Do not include explanations, just output the prompt."
40 ]
41 },
42 {
43 "id": 3,
44 "type": "OpenAIChatNode",
45 "pos": [
46 -5810,
47 2080
48 ],
49 "size": [
50 400,
51 580
52 ],
53 "flags": [],
54 "order": 2,
55 "mode": 0,
56 "showAdvanced": false,
57 "inputs": [
58 {
59 "name": "images",
60 "shape": 7,
61 "type": "IMAGE",
62 "link": 4
63 },
64 {
65 "name": "files",
66 "shape": 7,
67 "type": "OPENAI_INPUT_FILES",
68 "link": null
69 },
70 {
71 "name": "advanced_options",
72 "shape": 7,
73 "type": "OPENAI_CHAT_CONFIG",
74 "link": 3
75 }
76 ],
77 "outputs": [
78 {
79 "name": "STRING",
80 "type": "STRING",
81 "links": [
82 15
83 ]
84 }
85 ],
86 "properties": {
87 "Node name for S&R": "OpenAIChatNode",
88 "cnr_id": "comfy-core",
89 "ver": "0.20.1"
90 },
91 "widgets_values": [
92 "Describe this image",
93 false,
94 "gpt-5.6-sol"
95 ],
96 "color": "#432",
97 "bgcolor": "#653"
98 },
99 {
100 "id": 5,
101 "type": "LoadImage",
102 "pos": [
103 -6230,
104 2080
105 ],
106 "size": [
107 370,
108 370
109 ],
110 "flags": [],
111 "order": 1,
112 "mode": 0,
113 "inputs": [],
114 "outputs": [
115 {
116 "name": "IMAGE",
117 "type": "IMAGE",
118 "links": [
119 4
120 ]
121 },
122 {
123 "name": "MASK",
124 "type": "MASK",
125 "links": null
126 }
127 ],
128 "properties": {
129 "Node name for S&R": "LoadImage",
130 "cnr_id": "comfy-core",
131 "ver": "0.20.1"
132 },
133 "widgets_values": [
134 "example.png",
135 "image"
136 ]
137 },
138 {
139 "id": 19,
140 "type": "SaveText",
141 "pos": [
142 -5370,
143 2080
144 ],
145 "size": [
146 580,
147 580
148 ],
149 "flags": [],
150 "order": 3,
151 "mode": 0,
152 "inputs": [
153 {
154 "name": "text",
155 "type": "STRING",
156 "link": 15
157 }
158 ],
159 "outputs": [
160 {
161 "name": "text",
162 "type": "STRING",
163 "links": null
164 }
165 ],
166 "properties": {
167 "Node name for S&R": "SaveText"
168 },
169 "widgets_values": [
170 "Text/GPT",
171 "txt"
172 ]
173 }
174 ],
175 "links": [
176 [
177 3,
178 4,
179 0,
180 3,
181 2,
182 "OPENAI_CHAT_CONFIG"
183 ],
184 [
185 4,
186 5,
187 0,
188 3,
189 0,
190 "IMAGE"
191 ],
192 [
193 15,
194 3,
195 0,
196 19,
197 0,
198 "STRING"
199 ]
200 ],
201 "groups": [],
202 "config": [],
203 "extra": {
204 "ds": {
205 "scale": 0.5088635018188082,
206 "offset": [
207 7170.8099540117355,
208 -1440.4334910413975
209 ]
210 },
211 "frontendVersion": "1.48.3",
212 "VHS_latentpreview": false,
213 "VHS_latentpreviewrate": 0,
214 "VHS_MetadataImage": true,
215 "VHS_KeepIntermediate": true
216 },
217 "version": 0.4
218}