Uncaught (in promise) Error: Final run has not been received

API
1

I am working with code cloned from the OpenAI Assistants API Quickstart github repository. Long responses are getting truncated, and I’m getting the following error in the browser’s JavaScript console:

Uncaught (in promise) Error: Final run has not been received
at 122-1746a1df99a60643.js:9:10330

What’s the best way to address this?

2

Receiving same error. Service seemed to work find yesterday. Wonder if this has anything to do with recent update a few days ago.

3

removing the max_prompt token field when creating runs fixed it for me.
Likely context tokens has exceeded whatever maximum value you set for it and the run is unable to complete. I had a bunch of incomplete runs on my thread as well which lead me to this conclusion.

Not sure if anyone can shed some light on this on the proper way to use the max_prompt_tokens field as my limit was about twice the amount of tokens as my instruction prompts.

Feels like maybe the real issue is in run scoped prompt token limit being less than thread scoped prompt token limit

4

Thanks!
I’m not setting max_prompt token field. I think it may be an issue with where I am hosting the app (Vercel). When I test locally it works fine. Will report back what I find out.

5

Instructing the Assistant to limit responses to ~200 tokens fixes the problem, but that’s a weak workaround.

Seems to be a timeout issue on the Run when executed through the production environment (Vercel). Still investigating.

6

OK. Turns out it was a timeout issue with Vercel - What can I do about Vercel Functions timing out?

Setting maxDuration = 60; (60 seconds maximum duration on free plan) fixed the issue

1 Like