ComfyUI Node
Eric Qwen Reference Describer (VL→Text)
A ComfyUI node in Eric/QwenImage with 10 inputs and 2 outputs.
Eric Qwen Reference Describer (VL→Text)
- ref_image_1
- ref_image_2
- ref_image_3
- ref_image_4
- combined_prompt
- descriptions_log
◄grounded_prompt►
◄vl_api_urlhttp://localhost:30000►
◄model_name►
◄description_focusauto►
◄synthesis_modesynthesize►
◄max_image_dim512►
CategoryEric/QwenImage
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| grounded_prompt | STRING | Grounded prompt from EricGenSearcherNode. | |
| ref_image_1 | IMAGE | Primary reference image. | |
| vl_api_url | STRING | http://localhost:30000 | URL of the SGLang VL server (Gen-Searcher-8B). This is the same server used by EricGenSearcherNode. Default: http://localhost:30000 |
| ref_image_2opt | IMAGE | Optional 2nd reference. | |
| ref_image_3opt | IMAGE | Optional 3rd reference. | |
| ref_image_4opt | IMAGE | Optional 4th reference. | |
| model_nameopt | STRING | VL model override. Leave blank to auto-detect. | |
| description_focusopt | COMBO | auto | What to focus on when describing reference images: auto: let the VL model decide what's most important (recommended) full_scene: subject + clothing + setting + lighting subject_appearance: face, hair, build only clothing_and_style: garments and accessories only setting_and_environment: location and lighting only |
| synthesis_modeopt | COMBO | synthesize | synthesize (recommended): ask the VL model to weave the descriptions into the grounded prompt as a single paragraph. Produces the most natural and coherent result. append: simply append the descriptions to the grounded prompt. Faster (one fewer API call) but less coherent. |
| max_image_dimopt | INT | 512256–1024 | Max dimension when sending reference images to the VL model. 512px is sufficient for the model to extract visual details. Higher = slower API calls, diminishing returns. |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| combined_prompt | STRING | — |
| descriptions_log | STRING | — |