ComfyUI Node
Labnana Image Generation
Generate or edit images from a prompt (plus optional reference images). Handles submission, waiting and download internally; works for all sizes including 4K.
Labnana Image Generation
- client
- reference_images
- images
- image_urls
- task_id
◄prompt►
◄modelgemini-3-pro-image►
◄image_size2K►
◄aspect_ratio1:1►
◄seed0►
◄system_prompt►
◄reference_image_urls►
◄timeout600►
CategoryLabnana/Generate
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| client | LABNANA_CLIENT | — | |
| prompt | STRING | Generation or editing instructions. | |
| model | COMBO | gemini-3-pro-image | 6 options: gemini-3-pro-image, gemini-3.1-flash-image, gpt-image-2, wan2.7-image-pro, wan2.7-image, seedream-5-0-pro |
| image_size | COMBO | 2K | 4K is unsupported on wan2.7-image / seedream-5-0-pro and on reference-guided wan2.7-image-pro requests. |
| aspect_ratio | COMBO | 1:1 | Supported ratios depend on the selected model; invalid combinations fail locally. |
| seed | INT | 00–18446744073709550000 | Not sent to the API — changes force ComfyUI to re-run the node instead of using the cached result. |
| system_promptopt | STRING | Optional style/behavior instructions prepended to the prompt (the API has no native system-prompt field). | |
| reference_imagesopt | IMAGE | Batch of reference images, sent inline as base64. Max count depends on model (4-14). | |
| reference_image_urlsopt | STRING | Optional http(s):// or gs:// image URLs, one per line, sent as fileData references. | |
| timeoutopt | FLOAT | 60030–3600 | Max seconds to wait for the result (4K jobs can be slow). |
Outputs (3)
| Name | Type | Description |
|---|---|---|
| images | IMAGE | — |
| image_urls | STRING | — |
| task_id | STRING | — |