When an API call fails, the first question is always "why?". Our API Errors dashboard gives you the answer instantly. It automatically monitors and categorises different types of errors over time, helping you understand the reliability of your connected models.
This section breaks down issues into:
- Failed Requests: A general overview of all unsuccessful calls.
- Invalid Request Errors: Problems with your request format, like incorrect parameters.
- Rate Limit Errors: When you exceed the allowed request frequency for a model.
- Server Errors: Issues on the LLM provider's end, such as outages or system failures.
By understanding the source of errors, you can debug your code more effectively, manage your request pacing, and make informed decisions about which models are reliable enough for your production environment.