ComfyUI Node
VQGAN-CLIP Generator
A ComfyUI node in VQGAN-CLIP with 10 inputs and 1 output.
VQGAN-CLIP Generator
- init_image
- IMAGE
◄text_promptA painting of an apple in a fruit bowl►
◄iterations100►
◄seed0►
◄width512►
◄height512►
◄learning_rate0.10►
◄cutn32►
◄vqgan_model▾►
◄clip_model▾►
CategoryVQGAN-CLIP
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| text_prompt | STRING | A painting of an apple in a fruit bowl | — |
| iterations | INT | 1001–10000 | — |
| seed | INT | 00–18446744073709550000 | — |
| width | INT | 51264–2048 | — |
| height | INT | 51264–2048 | — |
| learning_rate | FLOAT | 0.100.001–1 | — |
| cutn | INT | 321–512 | — |
| vqgan_model | COMBO | 6 options: vqgan_imagenet_f16_16384, vqgan_imagenet_f16_1024, vqgan_gumbel_f8_8192, faceshq, wikiart_16384, sflckr | |
| clip_model | COMBO | 3 options: ViT-B/32, ViT-B/16, ViT-L/14 | |
| init_imageopt | IMAGE | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| IMAGE | IMAGE | — |