ComfyUI Node
DownloadAndLoadQWEN2_5_OMNIModel
A ComfyUI node in Qwen2_5-Omni with 4 inputs and 1 output.
DownloadAndLoadQWEN2_5_OMNIModel
- QWEN2_5_OMNI_model
◄modelQwen/Qwen2.5-Omni-3B►
◄quantization8bit►
◄attentionsdpa►
◄use_audio_outputtrue►
CategoryQwen2_5-Omni
Inputs (4)
| Name | Type | Default | Description |
|---|---|---|---|
| model | COMBO | Qwen/Qwen2.5-Omni-3B | 2 options: Qwen/Qwen2.5-Omni-7B, Qwen/Qwen2.5-Omni-3B |
| quantization | COMBO | 8bit | 3 options: none, 4bit, 8bit |
| attention | COMBO | sdpa | 3 options: flash_attention_2, sdpa, eager |
| use_audio_output | BOOLEAN | true | The model supports both text and audio outputs, if users do not need audio outputs, This option will save about 2GB of GPU memory but the return_audio option for generate function will only allow to be set at False.该模型支持文本和音频输出，如果用户不需要音频输出，此选项将节省大约 2GB 的 GPU 内存，但生成函数的 return_audio 选项只允许设置为 False。 |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| QWEN2_5_OMNI_model | QWEN2_5_OMNI_MODEL | — |