VercelDeployAI APIs

Vercel: FUNCTION_INVOCATION_TIMEOUT / 504 on AI requests

504: GATEWAY_TIMEOUT  Code: FUNCTION_INVOCATION_TIMEOUT

In short: Your serverless function ran longer than its maximum duration, which is common when waiting for a long LLM response.

Why it happens

Serverless functions have a max duration set by your plan and config. Waiting for a full non-streamed LLM completion, slow database queries, or chained API calls can exceed it, especially on cold starts.

How to fix it

1. Stream LLM responses to the client instead of waiting for the full completion.
2. Raise the function's max duration within your plan's limit (in Next.js: export const maxDuration = 60; in the route; check Vercel's current docs for your plan's ceiling).
3. Move long jobs to a background queue or cron and poll for the result.
4. Add timing logs to find which step is slow.

Copy-paste fix prompt

I'm getting this error in my app:

504: GATEWAY_TIMEOUT  Code: FUNCTION_INVOCATION_TIMEOUT

[Paste the full error, stack trace, and the file it points to here]

Likely cause: Your serverless function ran longer than its maximum duration, which is common when waiting for a long LLM response.

Task: Find the slow server route, add timing logs, switch the LLM call to streaming, and set an appropriate maxDuration. Suggest a background job if it would still be too slow.

Rules:
- First explain the root cause in 2 sentences and list the files you'll change.
- Make the smallest change that fixes it. Don't refactor unrelated code.
- Don't disable security or checks (RLS, auth, signature verification, type checking, CORS allow-all) to make the error disappear.
- Never print or hard-code secret keys.
- Finish by telling me exactly how to verify the fix.

Common with: v0, Cursor, Bolt. Last reviewed 2026-11-13.