O1-pro performance degradation (feb 8th) [possibily requires oai staff attention]

ChatGPT
1

this happens from time to time.

this time its o1-pro that struggling and acting like o1-mini…

started happening maybe 5-10 minutes ago

@vb @Foxalabs this is very similar to feb 5th with o3-mini (high and low)

probably requires oai staff attention

1 Like
2

its outputing maybe 1/6 of what it usually outputs

the quality went way down

its streaming responses, which is not usual for o1-pro

I maybe sent 10 messages to it in the last 3 hours

not what I expect from o1-pro

it would be nice to have someone take a look at it

and nothing from the status.openai.com:

The image displays a status dashboard showing various systems, including API, ChatGPT, Sora, Playground, and Labs, all marked as operational with high uptime percentages over the past 90 days. (Captioned by AI)
The image displays a status dashboard showing various systems, including API, ChatGPT, Sora, Playground, and Labs, all marked as operational with high uptime percentages over the past 90 days. (Captioned by AI)1928×1704 153 KB

but I’m glad that the feb 5th degradation event was recorded on the status page:

The image shows a status update log from February 5, 2025, detailing resolved and ongoing incidents involving increased errors for ChatGPT, Sora, and API, as well as batch creation errors. (Captioned by AI)
The image shows a status update log from February 5, 2025, detailing resolved and ongoing incidents involving increased errors for ChatGPT, Sora, and API, as well as batch creation errors. (Captioned by AI)2236×1232 218 KB

as at the time the issue started happening there was nothing up on the status page and multiple users were able to replicate the issue

here’s to hoping oai staff takes a look at this soon, here’s to trusting in the talent density at oai

1 Like
3

Same here, the thinking square isn’t appearing. I thought I had hit the hidden limit of O1-Pro, but it seems like it’s OpenAI’s issue now…

Quick fix: switch to mobile network

Similar issue at:
https://community.openai.com/t/deep-research-on-pro-account-broken-o1-and-o1-pro-in-web-version-answers-via-o3-mini-while-selecting-o1/1111339

1 Like
4

same thing happened to o3-mini 3 days ago, as far as I’m aware its not a limit but an issue. and the status page took hours to update. Not sure how long it took to get back to working, but it seems that it usually takes a couple of hours to get fixed

5

The same problem, in the pro version, the square doesn’t appear when the model is thinking. It started yesterday. However, it works in the mobile version. I logged in from another browser, a few requests worked, then again in o1-pro mode, some other model is thinking, obviously with weak responses.