ChatGPT Block

The ChatGPT Block allows you to seamlessly integrate OpenAI's language models into your API endpoint flows. It supports generating text responses based on a provided prompt, maintaining conversation history, and fine-tuning the model's behavior through advanced settings.

Configuration

Field	Required	Description
API Key	✅	Your OpenAI API key (usually mapped from the Secret Vault).
Model	✅	The specific OpenAI model to use (e.g., `gpt-4o`, `gpt-3.5-turbo`).
User Prompt	✅	The main text or question to send to the model.
System Prompt	—	Instructions that set the behavior and persona of the assistant.
Enable History	—	Toggle to maintain context across multiple requests.

History Settings

If Enable History is turned on, the following fields become active:

Storage Object Variable (Required): A variable mapped to a Storage Object to persist the chat history.
Chat ID: A unique identifier for the conversation session. If empty, a random ID is generated.
Message Count: The number of previous messages to include as context for the next request.

Advanced Settings

Toggle Advanced Settings to customize the model's generation:

Temperature (0 to 2): Controls randomness. Lower values are more deterministic; higher values are more creative.
Max Output Tokens: The maximum length of the generated response.
Top P (0 to 1): Nucleus sampling parameter.
Frequency Penalty (-2 to 2): Decreases the likelihood of the model repeating the same lines verbatim.
Presence Penalty (-2 to 2): Increases the likelihood of the model talking about new topics.

Output Variable

Specify a Variable Name to store the result.

Success Payload

{
  "success": true,
{
  "error_key": "rate_limit_exceeded",
  "error_message": "Too many requests sent in a short time. Please slow down."
}

Error Handling

API_KEY_MISSING / SECRET_ERROR — Missing or invalid API key.
MODEL_MISSING — No model was specified.
PROMPT_MISSING — The user prompt was empty.
HISTORY_VAR_MISSING — History was enabled but no storage object was provided.
invalid_api_key / insufficient_quota / rate_limit_exceeded — Standard OpenAI API errors.
RUNTIME_ERROR — An unexpected issue occurred during the API call.

ChatGPT