ComfyUI Node
Gemini Text Generation
A ComfyUI node in API Toolkit/Gemini/Text with 13 inputs and 1 output.
Gemini Text Generation
- text
◄api_key►
◄modelgemini-2.5-flash►
◄custom_model►
◄prompt►
◄system_instruction►
◄temperature0.70►
◄top_p0.95►
◄top_k0►
◄max_output_tokens0►
◄thinking_levelNONE►
◄thinking_budget0►
◄seed-1►
◄safety_settings_json►
CategoryAPI Toolkit/Gemini/Text
Inputs (13)
| Name | Type | Default | Description |
|---|---|---|---|
| api_key | STRING | Gemini API key. Leave blank to use GEMINI_API_KEY env var. | |
| model | COMBO | gemini-2.5-flash | Gemini model for text generation. |
| custom_model | STRING | Override with a custom model ID. | |
| prompt | STRING | The user prompt to send to the model. | |
| system_instructionopt | STRING | System instruction to guide model behavior. | |
| temperatureopt | FLOAT | 0.700–2 | Controls randomness. Lower = more deterministic. |
| top_popt | FLOAT | 0.950–1 | Nucleus sampling probability cutoff. |
| top_kopt | INT | 00–1000 | Top-K sampling (0 = disabled). |
| max_output_tokensopt | INT | 00–65536 | Max tokens in response (0 = model default). |
| thinking_levelopt | COMBO | NONE | How much the model should reason before answering. |
| thinking_budgetopt | INT | 00–100000 | Max thinking tokens (0 = model default). |
| seedopt | INT | -1-1–2147483647 | Random seed for reproducibility (-1 = random). |
| safety_settings_jsonopt | STRING | JSON safety settings from Safety Settings node. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |