[resolved] Is gpt3.5 API Timing Out a Known Issue?

[howon]

Is there a known issue for gpt3.5-turbo model? I’m using the NodeJS library and it always times out. This is how I’m calling the API.

  const response = await openai.createChatCompletion({
    model: 'gpt-4',
    messages: [{ role: 'user', content: "some prompt" }],
    max_tokens: 50,
    n: 1,
    stop: null,
    temperature: 1,

[ruby_coder]

Hi @howon

Yes, it’s very slow right now and I’m seeing the same error:

undefined method `key?' for "Error: Net::ReadTimeout Params:

HTH

[kennethzhang]

Same for GPT-4 now, I am using the early access API and it always times out after only one request.

I have also noticed that on the API account Usage page, the info cannot be loaded correctly after a long time.

It seems that there are some issues with their backend, perhaps?

[doomlaser]

Same here. There must be an outage at the moment. platform.openai.com doesn’t load properly, and I can’t view my usage.

[ruby_coder]

@howon

s there a known issue for gpt3.5-turbo model?

Yes, this is a known (current) API major outage issue:

https://status.openai.com

[ruby_coder]

FYI - Resolved: