ComfyUI Node
Qwen2_5_VL_Run_Advanced
A ComfyUI node in Qwen2_5-VL with 12 inputs and 1 output.
Qwen2_5_VL_Run_Advanced
- Qwen2_5_VL_model
- image
- video
- BatchImage
- text
◄system_text►
◄text►
◄video_decode_methodtorchvision►
◄max_new_tokens128►
◄min_pixels256►
◄max_pixels1280►
◄total_pixels20480►
◄seed1►
CategoryQwen2_5-VL
Inputs (12)
| Name | Type | Default | Description |
|---|---|---|---|
| system_text | STRING | — | |
| text | STRING | — | |
| Qwen2_5_VL_model | QWEN2_5_VL_MODEL | — | |
| video_decode_method | COMBO | torchvision | 3 options: torchvision, decord, torchcodec |
| max_new_tokens | INT | 1281–1024 | — |
| min_pixels | INT | 25664–1280 | Define min_pixels and max_pixels: Images will be resized to maintain their aspect ratio within the range of min_pixels and max_pixels. |
| max_pixels | INT | 128064–2048 | Define min_pixels and max_pixels: Images will be resized to maintain their aspect ratio within the range of min_pixels and max_pixels. |
| total_pixels | INT | 204801–24576 | We recommend setting appropriate values for the min_pixels and max_pixels parameters based on available GPU memory and the specific application scenario to restrict the resolution of individual frames in the video. Alternatively, you can use the total_pixels parameter to limit the total number of tokens in the video (it is recommended to set this value below 24576 * 28 * 28 to avoid excessively long input sequences). For more details on parameter usage and processing logic, please refer to the fetch_video function in qwen_vl_utils/vision_process.py. |
| seed | INT | 11–18446744073709550000 | — |
| imageopt | IMAGE | — | |
| videoopt | VIDEO | — | |
| BatchImageopt | BatchImage | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |