ComfyUI Node
Qwen3VL (External llama.cpp)
A ComfyUI node in Mira/SubPack/Qwen3VL with 8 inputs and 2 outputs.
Qwen3VL (External llama.cpp)
- image
- caption
- prompt
◄caption_typeDetailed Description►
◄caption_lengthlong►
◄extra_optionsUse single-line output description.►
◄temperature0.70►
◄top_p0.90►
◄max_tokens512►
◄llama_urlhttp://127.0.0.1:58080/v1/chat/completions►
CategoryMira/SubPack/Qwen3VL
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | — | |
| caption_type | COMBO | Detailed Description | 8 options: Detailed Description, Detailed Description (Casual), Straightforward Caption, Stable Diffusion Prompt, Tags (Danbooru style), Art Critique, +2 |
| caption_length | COMBO | long | 31 options: any, very short, short, medium-length, long, very long, +25 |
| extra_options | STRING | Use single-line output description. | Additional instructions to customize the caption. You can use this to add specific requirements or constraints for the caption generation. For example: Specify the style, focus, or any other details you want the caption to include. |
| temperature | FLOAT | 0.700–2 | Higher = more random, lower = more deterministic |
| top_p | FLOAT | 0.900–1 | Nucleus sampling parameter |
| max_tokens | INT | 5121–2048 | Maximum number of tokens to generate |
| llama_url | STRING | http://127.0.0.1:58080/v1/chat/completions | External llama.cpp service URL |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| caption | STRING | Generated image caption |
| prompt | STRING | Prompt used for caption generation |