o1 Pro unable to process over 35K tokens

[low pulsar]

Reported by @wispy nest

Bug Report: o1 Pro unable to process over 35K tokens

`Steps to Reproduce`

Something is very very wrong with o1 pro today.. Im not sure how you will reproduce this issue

`Expected Result`

1. Answers that normally take 2-5 mintes take 10-15 seconds
2. Im unable to even input prompts with 38K tokens ..i keep getting "The message you submitted was too long, please reload the conversation and submit something shorter."

I used 50-110K prompts multiple times last week and even a few days ago with zero issues ... i thought it was 128K context size?

3. I take the exact same 38K token prompt to o1 (not pro) and it works fine

`Actual Result`

The actual error i get is "The message you submitted was too long, please reload the conversation and submit something shorter." the same exact prompt / 38K also works fine with 4o, Deepseek, Claude and Gemini with Zero issues .. The same thing was happening a few weeks ago and then it just started working agin on its own.is there some sort of tracker that shows if a model is having issue that particular time?

Can OpenAI engineers please press restart somewhere lol ?

`Environment`

Using Firefox 135.0b4 (aarch64) Mac M1 Max

Jan 15, 2025, 2:36 PM

Additional Information

Please provide relevant details to help resolve the issue, such as:

• ChatGPT Shared Link (if applicable).
• Screenshots or videos demonstrating the problem.

-# ➜ Need to contact support? Visit the OpenAI Help Center.

[wispy nest]

i dont have a shared link but here is a screenshot of the error i keep getting

[feral jackal]

this should not happen even in plus role

[obtuse edge]

I am having same issue, anyone can help?

[strange steppe]

same here just happened today

[obtuse edge]

actually I cannot process message over 5k tokens