ComfyUI Node
vergora 图像编辑（图生图）
A ComfyUI node in vergora with 16 inputs and 3 outputs.
vergora 图像编辑（图生图）
- image
- mask
- image
- image_path
- image_url
◄prompt►
◄modelgpt-image-2►
◄size1024x1024►
◄image_url►
◄extra_json►
◄model_override►
◄qualityauto►
◄backgroundauto►
◄output_formatauto►
◄output_compression0►
◄moderationauto►
◄input_fidelitydefault►
◄n1►
◄user►
Categoryvergora
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| prompt | STRING | — | |
| model | COMBO | gpt-image-2 | 3 options: gpt-image-2, gpt-image-2.5-flare, gpt-image-2.5-sunburst |
| size | COMBO | 1024x1024 | 5 options: 1024x1024, 1536x1024, 1024x1536, 1080x1920, auto |
| imageopt | IMAGE | Source picture; a batch of up to 10 is sent as references | |
| image_urlopt | STRING | Picture URLs or data URIs, one per line; used when no IMAGE is connected | |
| extra_jsonopt | STRING | Extra request fields, merged into the body last | |
| model_overrideopt | STRING | Any other model id, wins over the dropdown | |
| maskopt | IMAGE | Optional edit mask; the transparent area is what gets repainted | |
| qualityopt | COMBO | auto | Official quality field; auto = do not send it |
| backgroundopt | COMBO | auto | auto / transparent / opaque |
| output_formatopt | COMBO | auto | auto / png / jpeg / webp |
| output_compressionopt | INT | 00–100 | 0-100, only sent for jpeg/webp (0 = do not send it) |
| moderationopt | COMBO | auto | auto / low |
| input_fidelityopt | COMBO | default | default / low / high - how closely the input is followed |
| nopt | INT | 11–10 | How many pictures to ask for; the gateway returned a single one in our tests |
| useropt | STRING | Official abuse-tracking field, sent when you fill it in |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| image | IMAGE | — |
| image_path | STRING | — |
| image_url | STRING | — |