Stream disconnected before completion due to TPM rate limit (250k TPM exceeded)
Resolved 💬 7 comments Opened Jan 28, 2026 by patricksuo Closed Jan 28, 2026
💡 Likely answer: A maintainer (github-actions[bot], contributor)
responded on this thread — see the highlighted reply below.
What version of Codex is running?
v0.92.0
What subscription do you have?
Pro
Which model were you using?
GPT5.2
What platform is your computer?
Darwin 24.6.0 arm64 arm
What terminal emulator and version are you using (if applicable)?
iTerm2
What issue are you seeing?
└ Stream disconnected before completion: Rate limit reached for organization org-BOvpEHVcDPTe8h4lZnwMO5Ly on tokens per min (TPM): Limit 250000, Used 250000, Requested 58487.
Please try again in 14.036s. Visit https://platform.openai.com/account/rate-limits to learn more.
What steps can reproduce the bug?
Uploaded thread: 019c02ac-dad9-74d1-a736-1748977e5a4a
What is the expected behavior?
_No response_
Additional information
Additional context: this is not caused by the 5-hour weekly usage limit.
At the time the error occurred, my weekly 5-hour limit had not been reached.
The stream was disconnected solely due to the TPM (tokens-per-minute) rate limit, which makes the failure confusing and difficult to anticipate from a user perspective.
7 Comments
Potential duplicates detected. Please review them and close your issue if it is a duplicate.
Powered by Codex Action
Uploaded thread: 019c0267-3efc-7ff2-a8ec-029cbedb2f21
Same. Tried re-login, didn't help. using v0.92.
Same. thread ID 019c0241-2e43-7b23-b8c7-544e0f395121
Same and there are now multiple tickets opened on the same bug
Fyi, it also started to charge you https://platform.openai.com/settings/organization/limits
I am guessing somehow your subscription is not considered and it things you are trying to use your API key; i tried to relog with my pro subscription, didn't help
It's working for me now. I didn't change anything.
Thanks for reporting. There was a temporary service issue that caused the responses API to incorrectly report rate limit errors when it shouldn't. The issue has been mitigated. Everything should be back to normal now. Apologies for the disruption.