AI prompt
AI Operation
AI Prompt is a widget flow node that allows you to synchronously invoke a GPT model with a custom prompt.
Description
You can use this for various data transformations, summarizations, or unique, non-repetitive experiences within your flow.
Configuration
Prompt
Prompt – a natural language prompt. You may reference context values within the prompt. The maximum allowed prompt length, including all context values, is 4096 characters. If this limit is exceeded, the widget flow will terminate with an error.
Temperature
Temperature – controls the randomness of the model's output. Accepts a value between 0 and 2. Lower values (e.g., 0.2) result in more deterministic and focused responses, while higher values (e.g., 1.5) produce more diverse and creative outputs.
Limitations
| Name | Value |
|---|---|
| Min temperature | 0 |
| Max temperature | 2 |
| Max output tokens | 512 |
| Max prompt input length | 4096 |
| Max compiled prompt length | 4096 |
| Inputs | Exactly one |
| Outpus | Exactly one |
Context values
An output field will be added to the context values, containing the full response from the model.