I created a few codex tasks and my 5 hour limit was full, I bought $40 credits since it was urgent and after few prompts more they were gone as well.
Now my limits are used for the entire week despite me buying a 1000 credits (which I honestly assumed would last the entire month).
Reddit was full of people complaining about this as well.
I’m hoping this is just an accident but some transparency or a quick fix would definitely be in order.
Thanks.
Hi - apologies for the disruption here. I believe this was caused by an issue on our end. We just shared an update here with this message:
"We’ve fixed a number of issues with tracking Codex cloud task usage, with more efficiency improvements for cloud tasks landing soon.
To thank you for your patience, we’ve granted $200 in free credits to Plus and Pro users who used cloud tasks in the past month, valid until Nov."
Please do let us know if you're still running into any issues.
It continues to malfunction, and no restoration or compensation credits have been provided.
I had 3,500 credits literally an hour ago and they’re all gone!!! Please return these credits to this account.
Hi everyone,
If you're still missing any credits, please write into support@openai.com so we can dig deeper into your account. If you're not able to get resolution there, let me know here and I'll follow up!
Hi this bug seems to be happening again. I’m using codex cli with gpt-5.2-codex reasoning medium, and I have burnt over 40dollars in half a day. Before the new year I never managed to spent my entire weekly limit on pro, now it’s gone within 3 days and 40dollers of credit is gone within a couple of hours.
Can confirm as well, pretty much all tasks fail and still it keeps taking my credits, down 40 USD in half a day
So i switched to the API and my reported usage today has been 206 requests and my context window (in vscode) shows 58% full, 149k tokens, so quick napkin maths shows 3million tokens input (206*149k) upper limit. but the api dashboard shows 18.5million tokens used…
Some more facts on the pricing using my API token usage breakdown:
for 06-11-2025 (6th of November 2025) the usage figures are:
model: gpt-5-2025-08-07
input_tokens total: 104448570
output_tokens total: 330191
input_cached_tokens total: 98334720
input_uncached_tokens total: 6113850
input_text_tokens total: 6113850
output_text_tokens total: 330191
input_cached_text_tokens total: 98334720
input price: $1.25/1000000 tokens
cached input token price: $0.125/1000000 tokens
output token price: $10.00/1000000 tokens
total price for uncached input: $7.6423125
total price for cached input: $12.29184
total price for output: $3.30191
total price: $23.2360625
The reported price for 2026-01-07 is: $51.303379, but it should only be $23.2360625. It’s an error of $28.0673165.
For 07-01-2026 (7th of January 2026) the usage figures are:
model: gpt-5.1-codex-max
input_tokens total: 18182412
output_tokens total: 53578
input_cached_tokens total: 15614080
input_uncached_tokens total: 2568332
input_text_tokens total: 2568332
output_text_tokens total: 53578
input_cached_text_tokens total: 15614080
total price for uncached input: $3.210415
total price for cached input: $1.95176
total price for output: $0.53578
total price: $5.697955
The reported price for 2026-01-07 is: $43.846953, but it should only be $5.697955. It’s an error of $37.788998.