AI request failed
For the complete documentation index, see llms.txt. For a full content snapshot, see llms-full.txt. Append .md to any kestra.io/docs/* URL for plain Markdown.

An AI Copilot request could not be completed.

  • HTTP status — 422 Unprocessable Content
  • type — https://kestra.io/docs/api-reference/problems/ai-request-failed

When you get it

  • The configured model provider rejects the call or returns an unusable response.
  • The prompt, or the flow it was asked to work on, cannot be processed.

Example response

A response reporting this problem:

{
  "type": "https://kestra.io/docs/api-reference/problems/ai-request-failed",
  "title": "AI request failed",
  "status": 422,
  "detail": "The model provider rejected the request: context length exceeded.",
  "instance": "/api/v1/main/ai/generate/flow"
}

How to handle it

This is a client error rather than a server error because the cause is usually the request or the provider configuration, not a fault in Kestra. Retrying an identical prompt rarely helps.

For the members every problem document carries, see Problem Details.

Was this page helpful?