ComfyUI Node
CLIP Generate Text (Image List)
Extends ComfyUI Generate Text with a system role and one-call IMAGE LIST support.
CLIP Generate Text (Image List)
- clip
- images
- video
- audio
- generated_text
◄system►
◄prompt►
◄model_formatauto►
◄max_length512►
◄sampling_mode▾►
◄thinkingfalse►
◄use_default_templatetrue►
CategoryOllama/CLIP
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| clip | CLIP | — | |
| system | STRING | — | |
| prompt | STRING | — | |
| model_format | COMBO | auto | Auto-detect the tokenizer, or manually select its chat format. |
| max_length | INT | 5121–32768 | — |
| sampling_mode | COMBO | 2 options: [object Object], [object Object] | |
| imagesopt | IMAGE | — | |
| videoopt | IMAGE | Video frames as one IMAGE batch; assumed to be 24 FPS. | |
| audioopt | AUDIO | — | |
| thinkingopt | BOOLEAN | false | Operate in thinking mode if the model supports it. |
| use_default_templateopt | BOOLEAN | true | Apply the model chat template. Disable only when prompt already contains the complete model-specific template. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| generated_text | STRING | — |