ComfyUI Node
Load CLIP with Visual
Load Qwen3-4B language model + Qwen3-VL visual encoder for Z-Image VL conditioning.
Load CLIP with Visual
- vl_model
◄lm_name▾►
◄visual_name▾►
Categoryloaders
Inputs (2)
| Name | Type | Default | Description |
|---|---|---|---|
| lm_name | COMBO | Qwen3-4B language model safetensors (qwen_3_4b.safetensors) | |
| visual_name | COMBO | Qwen3-VL visual encoder safetensors (qwen_3_vl_visual.safetensors) |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| vl_model | ZIMAGE_VL_MODEL | VLM model wrapper for CLIP Text Image Encode |