ComfyUI Node
jz OpenAI Image
A ComfyUI node in jz/api with 16 inputs and 4 outputs.
jz OpenAI Image
- image
- mask
- images
- mask
- cost
- usage
◄prompta photo of a cute dog►
◄modelgpt-image-2.5-sunburst►
◄size1024x1024►
◄custom_width2048►
◄custom_height2048►
◄qualityauto►
◄backgroundauto►
◄output_formatpng►
◄n1►
◄input_fidelityauto►
◄moderationauto►
◄output_compression100►
◄custom_model►
◄api_key►
Categoryjz/api
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | a photo of a cute dog | — |
| model | COMBO | gpt-image-2.5-sunburst | 4 options: gpt-image-2.5-sunburst, gpt-image-2.5-flare, gpt-image-2, custom |
| size | COMBO | 1024x1024 | auto lets the model pick. anything beyond the first three, and custom, needs gpt-image-2+ |
| custom_width | INT | 204816–3840 | — |
| custom_height | INT | 204816–3840 | — |
| quality | COMBO | auto | auto omits it. xhigh/max: gpt-image-2.5 only. cost scales steeply with quality |
| background | COMBO | auto | transparent needs png or webp; the alpha comes out on the mask output |
| output_format | COMBO | png | 3 options: png, jpeg, webp |
| n | INT | 11–10 | each image is billed |
| imageopt | IMAGE | wired = edit. every frame is one reference, max 16 | |
| maskopt | MASK | applies to the first image; 1 = area to change | |
| input_fidelityopt | COMBO | auto | edits only. auto omits it, which gpt-image-2+ requires |
| moderationopt | COMBO | auto | 2 options: auto, low |
| output_compressionopt | INT | 1000–100 | jpeg/webp only |
| custom_modelopt | STRING | — | |
| api_keyopt | STRING | — |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| images | IMAGE | — |
| mask | MASK | — |
| cost | STRING | — |
| usage | STRING | — |