ComfyUI Node
Image Captionator Qwen 3.5
A ComfyUI node in Captionator with 7 inputs and 2 outputs.
Image Captionator Qwen 3.5
- image
- caption
- full_output
◄model▾►
◄promptWrite a clear and detailed description of the given image in one concise paragraph (maximum 200 words). Focus on key visual elements such as main subjects, their appearance, positions, actions, environment, lighting, colors, mood, and any notable details. Avoid speculation or assumptions beyond what is visible. Use precise, descriptive language while keeping the text compact and well-structured.►
◄resize_to0►
◄max_new_tokens256►
◄seed0►
◄thinkfalse►
CategoryCaptionator
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | 3 options: [Download] Qwen 3.5 2B, [Download] Qwen 3.5 4B, [Download] Qwen 3.5 9B | |
| prompt | STRING | Write a clear and detailed description of the given image in one concise paragraph (maximum 200 words). Focus on key visual elements such as main subjects, their appearance, positions, actions, environment, lighting, colors, mood, and any notable details. Avoid speculation or assumptions beyond what is visible. Use precise, descriptive language while keeping the text compact and well-structured. | — |
| resize_to | INT | 00–4096 | — |
| max_new_tokens | INT | 2561–8192 | — |
| seed | INT | 00–9223372036854776000 | — |
| think | BOOLEAN | false | — |
| imageopt | IMAGE | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| caption | STRING | — |
| full_output | STRING | — |