ComfyUI Node
Qwen2 VQA
A ComfyUI node in Comfyui_Qwen2-VL-Instruct with 10 inputs and 1 output.
Qwen2 VQA
- source_path
- STRING
◄text►
◄modelQwen2-VL-2B-Instruct►
◄quantizationnone►
◄keep_model_loadedfalse►
◄temperature0.7►
◄max_new_tokens2048►
◄min_pixels200704►
◄max_pixels1003520►
◄seed-1►
CategoryComfyui_Qwen2-VL-Instruct
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| text | STRING | — | |
| model | COMBO | Qwen2-VL-2B-Instruct | 6 options: Qwen2-VL-2B-Instruct-GPTQ-Int4, Qwen2-VL-2B-Instruct-GPTQ-Int8, Qwen2-VL-2B-Instruct, Qwen2-VL-7B-Instruct-GPTQ-Int4, Qwen2-VL-7B-Instruct-GPTQ-Int8, Qwen2-VL-7B-Instruct |
| quantization | COMBO | none | 3 options: none, 4bit, 8bit |
| keep_model_loaded | BOOLEAN | false | — |
| temperature | FLOAT | 0.70–1 | — |
| max_new_tokens | INT | 2048128–2048 | — |
| min_pixels | INT | 2007043136–12845056 | — |
| max_pixels | INT | 10035203136–12845056 | — |
| seed | INT | -1 | — |
| source_pathopt | PATH | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| STRING | STRING | — |