Skip to main content

Error response format

Every public API error includes a stable code, a human-readable message, actionable guidance, and a request ID:
Handle error.code rather than matching the complete message. Store the request_id in logs and include it when contacting AtlasFlux support.

Retry policy

  • Do not retry 400, 401, 403, or 409 errors until the request, key, permission, or idempotency key is fixed.
  • Do not automatically retry 402 errors; add credits or adjust the spending limit first.
  • For 429, wait for retry_after_seconds or the Retry-After response header.
  • Retry 502, 503, and 504 with exponential backoff and a maximum retry count.
  • Retry a 500 response only when error.retryable is true.

JavaScript example

Context-limit errors

Context errors include measured token counts when AtlasFlux can calculate them. Reduce the conversation history, remove unnecessary content, or lower max_output_tokens; increasing the output limit does not fix a context overflow.

Billing on failed requests

Requests recorded with status failed are not charged. AtlasFlux releases the complete temporary reservation and records the request cost as RM0.000000, including when search or content extraction was attempted before the failure. Use the error request_id to locate the matching zero-cost entry in Logs.

Streaming errors

Streaming requests can fail after the HTTP connection has already returned 200. Always handle the stream error event and stop processing further output. See Streaming responses for event examples.