You can throttle requests in Clay by setting a Request Limit and Duration in the HTTP API settings. This controls how many requests can be sent within a specific timeframe.
To set this up:
1. Go to your HTTP API integration settings
2. Find the "Define rate limit" section
3. Enter your maximum number of requests in the "Request Limit" field
4. Set the time window in milliseconds in the "Duration (in ms)" field
For example, setting 10 requests with 1000ms duration would limit to 10 requests per second.
The rate limiting setting in Clay's HTTP API integration is general and not specific to OpenAI. You can set a Request Limit (number of requests) and Duration (in milliseconds) to control how many API requests are sent within a timeframe.
For OpenAI specifically, you'll need to ensure your OpenAI API key has sufficient token per minute (TPM) limits. For reliable use with Clay, you need 30, 000 TPM for ChatGPT columns and 450, 000 TPM for Claygent.
If you're experiencing slow performance with OpenAI, consider upgrading your OpenAI API key to a higher tier.
.png){kind=logo}
