ComfyUI Node
Gemini Prompt Refiner
A ComfyUI node in API Toolkit/Gemini/Text with 7 inputs and 1 output.
Gemini Prompt Refiner
- refined_prompt
◄api_key►
◄modelgemini-2.5-pro►
◄custom_model►
◄base_prompt►
◄system_instruction►
◄thinking_levelNONE►
◄temperature0.70►
CategoryAPI Toolkit/Gemini/Text
Inputs (7)
| Name | Type | Default | Description |
|---|---|---|---|
| api_key | STRING | Gemini API key. Leave blank to use GEMINI_API_KEY env var. | |
| model | COMBO | gemini-2.5-pro | Gemini model for prompt refinement. |
| custom_model | STRING | Override with a custom model ID. | |
| base_prompt | STRING | The prompt to refine and improve. | |
| system_instructionopt | STRING | Instructions for how to refine the prompt. Leave empty for default behavior. | |
| thinking_levelopt | COMBO | NONE | How much the model should reason before answering. |
| temperatureopt | FLOAT | 0.700–2 | Controls randomness in refinement. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| refined_prompt | STRING | — |