I get this error when making an API call to Davinci model 3. Given that the GPT completion is part of the front end user experience for my site, this is a critical error as it disables the site completely. What’s going on?
Status code 503
{
"error": {
"message": "That model is currently overloaded with other requests. You can retry your request, or contact us through our help center at help.openai.com if the error persists. (Please include the request ID 7ef84fe5909aa6768fb8d39a081a5ccc in your message.)",
"type": "server_error",
"param": null,
"code": null
}
}
If OpenAI can’t maintain the capacity for GPT-3 requests they really need to stop taking on new accounts until they can keep up or take capacity off of ChatGPT. It’s not fair that paid users don’t get a reliable service - there was another outage just 2 days ago.
I just came here to make a post about this, I’ll go ahead and tag along to this one.
I agree that it is unfair, we just scaled to $2,000/week in API calls and would really appreciate if our users didn’t have these outages.
Inside of our Facebook group we actually started getting a ton of posts over the last couple days about documents not being generated. I knew this was a problem sometimes, but no one ever posted about it that’s how rare it was.
I decide to take a look for myself and see the same 503 error code. And when it’s not the 503 error code, it’s a different one that I’m making too many request at once, which there’s no way is true.
Feeling… Frustrated, but not too upset because things are early. They do need to chill out on taking on more users though until they scale up their servers
And absolutely, you can utilize GPT to build some really powerful apps.
In terms of what I used to build mine, I use the React programming language, firebase for authentication, OpenAI for my AI of course and express.js for by backend. Highly recommend this stack if you are a beginner.
All of you that pay OpenAI for the use of any API should demand the company to protect your investment in the technology against those that dabble / play for free…
H.Rojas
All of you that pay OpenAI for the use of any API should demand the company to protect your investment in the technology against those that dabble / play for free…
H.Rojas
Another outage! Keep getting 503 errors and 429 Errors on the GPT-3 API which given that I’m in testing cannot possibly be correct. Anybody know anything about openai addressing these issues?
P.S. the status page indicates all systems are operational…
429’s and 503 are becoming way too common. We really need a dedicated, robust server for paying customers only! Customer support never replies. I reply to them, and their next email is “Are you satisfied with our help? Thank you!”
I really hope they aren’t using AI to reply to us, ironic, but extremely unhelpful.