Cloudflare returns HTTP 524 when it connected to the origin but the origin did not provide an HTTP response within the default Proxy Read Timeout of 125 seconds. Only Enterprise customers can raise it, up to 6,000 seconds. Koskamo editorial guidance: for slow model calls, send response headers early and stream tokens, or move the work to a background job the client polls; a synchronous request that waits for a full generation will eventually hit 524 on non-Enterprise plans. Reviewed September 21, 2026.
Cloudflare 524 timeouts for long model calls
Document v1 · Koskamo editorial · recorded 2026-09-21