ComfyUI Node
Imagen Image Generation
A ComfyUI node in NanoBanana2/Image with 11 inputs and 1 output.
Imagen Image Generation
- network
- images
◄api_key►
◄modelimagen-4.0-generate-001►
◄prompt►
◄custom_model►
◄number_of_images1►
◄aspect_ratio1:1►
◄negative_prompt►
◄seed-1►
◄safety_filter_levelblock_low_and_above►
◄person_generationallow_adult►
CategoryNanoBanana2/Image
Inputs (11)
| Name | Type | Default | Description |
|---|---|---|---|
| api_key | STRING | NanoBanana - API key. Leave blank to use GEMINI_API_KEY env var. | |
| model | COMBO | imagen-4.0-generate-001 | Imagen model. Ultra = highest quality, Standard = balanced, Fast = cheaper/faster. |
| prompt | STRING | Describe the image to generate. | |
| custom_modelopt | STRING | Override with a custom model ID (takes priority over dropdown). | |
| number_of_imagesopt | INT | 11–4 | How many images to generate in one call (Imagen supports up to 4). |
| aspect_ratioopt | COMBO | 1:1 | Output aspect ratio. Imagen supports: 1:1, 3:4, 4:3, 9:16, 16:9. |
| negative_promptopt | STRING | Things to avoid in the image. | |
| seedopt | INT | -1-1–2147483647 | Seed for reproducibility. -1 = random. (0 is a valid seed for some models — v2.0 silently dropped seed=0; v2.1 honors it. Use -1 for random.) |
| safety_filter_levelopt | COMBO | block_low_and_above | Safety filter threshold. Developer API often only accepts block_low_and_above on free tier; Vertex AI supports all four. API surfaces a clear error if the level isn't accepted for your model/account. |
| person_generationopt | COMBO | allow_adult | Whether/how to generate people. Some models enforce stricter defaults. |
| networkopt | NB_NETWORK | Optional. Wire a NanoBanana - Network Route node here to route this request through that proxy (e.g. US egress). |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| images | IMAGE | — |