Codex 28 · Theme · #9
Quota usage increases unexpectedly
44people describe this
44 on X
X replies to the promise post are counted apart from Rantr votes. “This happened to me too” adds your browser to this theme on Rantr, once.
Where OpenAI stands
Shipped · Day 2Day 13: Tibo announces a limit reset as the Day 13 delivery ↗Now If you're signed in with a ChatGPT account, turn on Auto-review in Settings > General > Permissions > Auto-review
In their words
What people said
- Problem
Something is wrong with Codex usage limits. I'm running the same routine tasks via cron as before - just splitting up the work and triggering automations. Nothing has fundamentally changed in my workflow, yet my usage limits are being consumed roughly 4 times faster than before.
@MiharuEdge on X ↗ · 25 Oct 2026
- Problem
Can you guys please fix the usage bug there’s no way 20 min of an Astra job should be taking 11 percent of my weekly usage
@khiazmos on X ↗ · 25 Oct 2026
- Problem
Thanks Tibo 🙏 Can the team please look into Dot possibly consuming usage in the background? I did not activate my Dot until 2 days ago—yes, I was late haha, needed to finish high priority coding tasks before learning new tools—and since trying a few prompts I noticed all other…
@Nebula_Matrix on X ↗ · 25 Oct 2026
- Mixed
Hi Tibo, after the update, I’ve noticed that the quota is being consumed much faster now. That said, the processing speed does feel noticeably quicker too. I’m not sure if this is expected behavior following this minor update, or if it’s another bug.
@chnsh188499 on X ↗ · 24 Oct 2026
- Problem
Can you tell me why when I use Dot, it hits my usage hard when I have it commit to GitHub for instance? That sure isn’t free usage. I guess only chat is free if you don’t do anything with it?
@Jeff_S_Hatton on X ↗ · 24 Oct 2026
- Problem
I started to reach limit on Sol 6.1 after devday. it seems like CLI is taking up tokens more
@nowgeun on X ↗ · 24 Oct 2026
- Problem
Strange higher consumption detected in gpt 6.1 Only high reasoning. Same job as yesterday, a few hours ago, before the reset. After the reset, it's really consuming a lot more
@FrancescoM51775 on X ↗ · 23 Oct 2026
- Problem
@OpenAI What’s the point of usage limit resets if the same tasks now consume several times more tokens after the October updates? Support keeps sending generic replies. I prefer GPT over Claude, but you are making it impossible to stay. This is ridiculous.
@MrDarkthefirst on X ↗ · 23 Oct 2026
- Problem
Always cost 25% my first 5h limit after reset with only one click (model: GPT 6.1 Sol - light). It that normal?
@oicufo on X ↗ · 23 Oct 2026
- Problem
One request used up my entire 5-hour GPT-6.1 Sol limit in 3 minutes, and Astra's in just 1 minute! What changed? Why am I paying if I can only send one request before waiting 5 hours? This is ridiculous and makes the service nearly unusable. I may really have to switch to Claude.
@PonceDeLeonnn on X ↗ · 21 Oct 2026
- Problem
これまでは10%の消費で済むと確認できているプロンプトで50%消費しましたが、私にはあなたからの連絡が来ていません。
@FanFunGOGO on X ↗ · 20 Oct 2026
- Problem
Bruh I think it’s worse, whole 5 hour usage got used in 1 single prompt.. been using GPT 6.1 Sol low, and I used to get at LEAST 2 hours usage, but 1 30 minute prompt? This is WORSE!
@ColdSparkUK on X ↗ · 20 Oct 2026
- Problem
@thsottiaux sir i think this particular change broke rate limits for everyone and is now draining everything, did you double check it?
@BarrellTitor44 on X ↗ · 19 Oct 2026
- Problem
Codex is broken how can I burn 5 hr limit with one task Import a FBX to unreal engine like a week ago I could ask the same task and be able to do it 100 times now 1 FBX like it's just sad.....
@stati83790 on X ↗ · 19 Oct 2026
- Problem
Same task, same prompt: Codex used my full 5-hour limit + 200 extra credits. Claude Code used only 13%. This is not reasonable. Keep this up and I’m leaving OpenAI.
@Willam445938 on X ↗ · 19 Oct 2026
- Problem
Me too. Oct 16, 2026, 18:09:36→18:10:02 UTC: Codex weekly usage jumped 31%→100%. Logs from my PC show one Sol response: 94K tokens, 83K cached. Credit deductions followed at 18:10:17. The quota record format also changed. Cause unconfirmed; reported to Support.
@Libratechw on X ↗ · 18 Oct 2026
- Problem
$200/month Pro. 5 hours, one window, one task: build a 3-agent backtest. 47% of my weekly allowance gone. Still not working. Two weeks ago, 3 windows running all day used ~20%. This is ridiculous. What changed? Where's the usage breakdown?
@ISSA0GRAVITY on X ↗ · 18 Oct 2026
- Mixed
So yesterday i couldnt burn my quota to save my life even on ultra. Today after this new reset im burning my usage at like 10x what i was yesterday.... What is going on tibo :(
@Schedul1 on X ↗ · 17 Oct 2026
- Idea
Tibo, I think you might want to look into why our quota is burning through so fast. I've noticed that gpt6.1sol is draining credits about four to five times faster than before.
@nonshan454109 on X ↗ · 17 Oct 2026
- Problem
Two Codex tasks working together on my backend used to consume just 10% of my weekly limit COMBINED on the $200 Pro plan. Yesterday, the same two tasks burned through 100% of that limit and started draining paid credits. How the hell is that an improvement?
@korkin_igor on X ↗ · 17 Oct 2026
- Problem
Man I burned 4 of my 500$ accounts in 6 hours, there is clearly something wrong. Check github for latest users complains please.. you cant say this while token burn is at this rate... honestly this is un-acceptable now
@khomakho on X ↗ · 17 Oct 2026
- Problem
There is a bug with Pro accounts where usage went straight to 0% earlier today and now it's using up credits. I was at ~70% weekly and have been at 0% since early afternoon. Other similar reports piling up.
@renatomoraesp on X ↗ · 17 Oct 2026
- Mixed
1. This feature was existed on day10 at least; 2. I feel it significantly increased the token usage 3. To make codex subagent work fresh with small task i now use Claude lead agent to orchestrate the implementation.
@GabrielChengtao on X ↗ · 17 Oct 2026
- Problem
Hi Tibo, since yesterday's update, my credit usage has spiked — it's burning through noticeably faster than before. Could you please look into this?
@Capricorn9815 on X ↗ · 17 Oct 2026
- Problem
I consumed 100% usage in 1 day with gpt6.1 sol light, something changed I think, since 2 days ago my usage is burning like crazy with simple tasks.
@JohnLiquidity on X ↗ · 17 Oct 2026
- Problem
Today i had 70% weekly usage left then it dropped to 0 for no reason. I have $100 pro subscrption.
@nimaa_mrd on X ↗ · 16 Oct 2026
- Win
Usage has increased a lot today. Full pro100 gone with two sol 6.1 low threads doing gpu coding.
@FundergroundDev on X ↗ · 15 Oct 2026
- Problem
Since yesterday, My GPT 6.1 Sol 5-hour usage limit seems to be getting consumed much faster than before. Previously, it used to last significantly longer. Is anyone else experiencing this since the 475K context window was added to GPT 6.1 Sol?
@bkrsna_ai on X ↗ · 15 Oct 2026
- Problem
whats the issue with usage. gpt 6.1 sol medium burns almost faster than it used to? what happend. one prompt and i spent all my limit. i used to never run out with same prompts!! happening last 3 days and today islts the worst
@Wibwick on X ↗ · 15 Oct 2026
- Problem
I’ve been a subscriber for a long time—did you quietly reduce the quota for the 20x tier? For the exact same project, it used to consume 20% of my quota for one hour of "fast" mode; now, it consumes 20% of my quota even without "fast" mode enabled.
@kangebutu on X ↗ · 15 Oct 2026
- Problem
So.... How come Astra on Medium basically just setting up a work email through gmail take up 3% of the weekly usage in 20 minutes? Just wondering. No fast usage, 3% in 20 minutes means At the same usage rate: Total: 11 hours, 6 minutes, 40 seconds to reach 100%. Remaining…
@AMerchantmoh on X ↗ · 13 Oct 2026
- Problem
Why did I only use GPT6.1Sol for 3 minutes today, and he didn't do anything, and directly consumed 56% of my five-hour quota? This is terrible. I don't know what the meaning of the 5-hour limit is, except for making users have a bad experience.
@JacksonChe3983 on X ↗ · 13 Oct 2026
- Mixed
How can people consume so much in 5 hours? When the limit used to be 20x, I never hit the 5-hour limit like, ever. I use 1.5B+ tokens every day. What kind of work are you guys doing?
@paxmanam on X ↗ · 11 Oct 2026
- Problem
What happened to Sol 6.1? I'm burning through 5-hour tokens in 40 minutes.
@mulekeproBR on X ↗ · 11 Oct 2026
- Problem
为什么我没有任何在跑的任务,只是在codereview里的pr里不停的 `@codex review`,却发现我的codex额度在变低,不是说不消耗额度的吗?
@johnsmithd37k on X ↗ · 10 Oct 2026
- Problem
Didnt recived it yesterday, didnt recived it today. on 100$ plan, and tokens are burning way faster since last update =S
@gruzrh on X ↗ · 9 Oct 2026
- Problem
@thsottiaux GPT 6.1 Sol Xhigh is using more credit limit then yesterday almost looks as if it is 2x the speed for consumption. Can you please take a look at it. I had Astra on for a short time then switched back which is when I noticed it was using more than before
@BilalRashi10543 on X ↗ · 8 Oct 2026
- Problem
Since the reset, token consumption has increased significantly. Is there a configuration setting affecting this?
@daisyxiaqiu on X ↗ · 8 Oct 2026
- Win
Ngl, now it feels like the speed of tokens disappearing was increased by like 100%.
@DavidKPaul on X ↗ · 8 Oct 2026
- Problem
So today gpt 6.1 Sol usage has gone back to what I was getting the first several days it was out. Can you confirm there was an issue that caused massive usage burn over a several day period? It seems odd that today everything was back to normal after several days of rapid usage…
@the_jimmy_jones on X ↗ · 7 Oct 2026
- Win
これヤバくない? Codexチームに作業させた時、まさにこの自動承認のトークン消費が無視できないレベルで高くてきつかったから、超助かる
@wemp12 on X ↗ · 6 Oct 2026
- Problem
Is Codex Pro 20x usage accounting broken? Oct 4: 224M tokens used 10% of weekly allowance; Oct 5: 201M used 26%. Remaining allowance fell from 70% to 21%, including 30%→21% in under an hour. Separate bug: switching to Sol Light left existing subagents on Astra low.
@Ryhrw4y on X ↗ · 5 Oct 2026
- Problem
Why is tokens usage becoming soo high? I finished my weekly limit in 1d although didn't increase usage ..
@wassimbs2 on X ↗ · 5 Oct 2026
- Problem
Don’t know about improvement, but apparently today saying "hi" to GPT-6 Luna Light in a brand new chat costs 1% of my weekly limit and 4% of my 5-hour quota.
@LooteraFriendly on X ↗ · 5 Oct 2026
Your experience
Had this too? Say what happened.
JEV reads your words as you type and shows which theme they belong to. It may place them in another theme if that fits better.
Status
Where OpenAI stands
“Shipped” means an announcement was linked to the theme by AI, not that the problem is fixed for everyone. “Not addressed yet” means no recorded announcement was linked; it is not proof that nothing changed.
- Reset · Day 13Day 13: Tibo announces a limit reset as the Day 13 delivery ↗
- Reset · Day 3Day 3: Tibo loads a banked reset in every paid account ↗
- Shipped · Day 2Day 2.4: Decisions API enters public beta ↗
- Shipped · Day 2Day 2.1: Auto-review is now free with ChatGPT sign-in ↗
Now If you're signed in with a ChatGPT account, turn on Auto-review in Settings > General > Permissions > Auto-review
- Reset · Day 2Day 2: Tibo says the reset has been processed after the community poll ↗
Corrections
Changes to this theme
-
“Tool actions count against usage” was merged into this theme: JEV judged them the same problem.
-
“Background usage consumption by Dot” was merged into this theme: JEV judged them the same problem.
-
“Fresh subagent context needed” was merged into this theme: JEV judged them the same problem.
-
“CLI uses more tokens” was merged into this theme: JEV judged them the same problem.
-
“Estimated usage differs from actual” was merged into this theme: JEV judged them the same problem.
-
“Rate limits drain after rollout” was merged into this theme: JEV judged them the same problem.
-
“Usage rate shifts after reset” was merged into this theme: JEV judged them the same problem.
-
“Weekly usage counter resets unexpectedly” was merged into this theme: JEV judged them the same problem.
-
Renamed from “Unexpected token usage increase” to “Quota usage increases unexpectedly” as more people described it.
-
“Greeting uses quota in new chats” was merged into this theme: JEV judged them the same problem.
-
“Usage window closes earlier” was merged into this theme: JEV judged them the same problem.
-
“Automatic approval token usage” was merged into this theme: JEV judged them the same problem.
-
“Review requests count toward usage” was merged into this theme: JEV judged them the same problem.
-
“Allowance updates in jumps” was merged into this theme: JEV judged them the same problem.
-
“Accelerated token drain” was merged into this theme: JEV judged them the same problem.
Share
Share this theme
Follow
Follow this theme by email
We send a confirmation link first; nothing else is sent until you confirm. Then at most one email a day, only when the theme changes.
X thread: 2769 replies to the promise post collected (4 Oct 2026 – 26 Oct 2026): 2555 placed in a theme and 214 set aside as off-topic or unsafe. X replies are found by search and each one is checked against its public post on X. Last read: 26 Oct 2026, 23:46 UTC. Not every reply is read.
← Every theme