ComfyUI Node
Gemini Detect
Detect objects in an image and emit their bounding boxes as regions for the Regional Prompt Builder.
Gemini Detect
- image
- client
- regions
◄objects►
◄seed-1►
◄modelgemini-3.5-flash►
◄temperature0.00►
◄max_objects20►
CategoryERPK/Gemini
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | Image to detect objects in (ComfyUI tensor) | |
| objects | STRING | Objects to detect, one per line (e.g. 'red car'). Leave empty to detect all prominent objects. | |
| seed | INT | -1-1–2147483647 | Seed for reproducible detection. Randomizes by default. |
| clientopt | GEMINI_API_CLIENT | Gemini API client from Gemini API Config node (optional if API key is configured in Settings) | |
| modelopt | COMBO | gemini-3.5-flash | Gemini model to use for detection |
| temperatureopt | FLOAT | 0.000–2 | Lower = more deterministic detection |
| max_objectsopt | INT | 201–100 | Maximum number of regions to return |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| regions | ERPK_REGIONS | Detected regions as JSON for the Regional Prompt Builder's regions input. |