#Headers Timeout (gpt-3.5-turbo)

1 messages · Page 1 of 1 (latest)

thin stream
#
  • /v1/chat/completions
    Getting HeadersTimeout Errors when using a rather longer prompt.

code: 'UND_ERR_HEADERS_TIMEOUT'

I am assuming this is a prompt length issue, as a short Prompt (no more than 10 words) has 100% success rate out of say 10 calls.

With a long prompt >10; I am seeing almost 100% error rate.

Another note - I cannot replicate the issue when using postman.

Tried calling endpoint via Axios, Fetch and Curl.