AI nodes fail after 5 minutes with UND_ERR_HEADERS_TIMEOUT

Any AI node calling a slow model (often a local one) fails once a response takes longer than five minutes.

Difficulty

Easy

Time

~5 min

Steps

Fixed in

n8n@2.5.2

Last checked

Oct 17, 2026

Verified fix: PR #24292 merged on 2026-01-15. Fixed in n8n@2.5.2 (first stable release containing the merge commit). Last checked 2026-10-17.

Symptom

Any AI node calling a slow model (often a local one) fails once a response takes longer than five minutes.

Evidence what you'll see

HeadersTimeoutError with code UND_ERR_HEADERS_TIMEOUT
Always at about the five-minute mark
Shorter prompts succeed

Cause

The HTTP client used by AI nodes kept its default 300-second header and body timeouts, which no setting could change.

Fix steps

0 of 2 done

Step 1Upgrade n8n to a release that includes PR #24292.
Step 2Set the N8N_AI_TIMEOUT environment variable (in milliseconds) if you need more than its one-hour default, then restart n8n.

Your check marks stay in this browser only.

Verify the fix worked

Run a workflow whose model call takes more than five minutes and confirm it finishes instead of timing out.

Sources and credit

Summarized in our own words from these public reports. Credit goes to their authors.

Provided as is, without warranty. Try changes in a non-production environment and back up first. Security-related steps: verify against the official docs. Terms · Request removal or attribution change

Hit an error we don't cover? Submit it