Copilot rate limit but i have PRO licence !!! · community · Discussion #180092

17 min read Original article ↗

💬 Your Product Feedback Has Been Submitted 🎉

Thank you for taking the time to share your insights with us! Your feedback is invaluable as we build a better GitHub experience for all our users.

Here's what you can expect moving forward ⏩

  • Your input will be carefully reviewed and cataloged by members of our product teams. 
    • Due to the high volume of submissions, we may not always be able to provide individual responses.
    • Rest assured, your feedback will help chart our course for product improvements.
  • Other users may engage with your post, sharing their own perspectives or experiences. 
  • GitHub staff may reach out for further clarification or insight. 
    • We may 'Answer' your discussion if there is a current solution, workaround, or roadmap/changelog post related to the feedback.

Where to look to see what's shipping 👀

  • Read the Changelog for real-time updates on the latest GitHub features, enhancements, and calls for feedback.
  • Explore our Product Roadmap, which details upcoming major releases and initiatives.

What you can do in the meantime 💻

  • Upvote and comment on other user feedback Discussions that resonate with you.
  • Add more information at any point! Useful details include: use cases, relevant labels, desired outcomes, and any accompanying screenshots.

As a member of the GitHub community, your participation is essential. While we can't promise that every suggestion will be implemented, we want to emphasize that your feedback is instrumental in guiding our decisions and priorities.

Thank you once again for your contribution to making GitHub even better! We're grateful for your ongoing support and collaboration in shaping the future of our platform. ⭐
You must be logged in to vote

37 replies

@advicee12

THIS IS A FUCKING JOKE, IM AT 34%, I HAD RATE LIMIT LAST WEEK AFTER 2 days, OF FEW REQUESTS, TODAY WAS THE RESET, I DID ONE REQUEST OF 5.5 GPT JUST TO GET RATE LIMITED, MY SECOND REQUEST STOPPED AFTER THAT BECAUSE OF ANOTHER FUC***** RATE LIMIT!!!! PRO+ HOW WHEN I CANT EVEN FU*** USE 2 REQUESTS, I PAY FOR HOW MANY REQUESTS??????? THIS IS A FU*&& SCAM, can't EVEN USE AUTO TO FINISH THE WORK THAT AGENT STARTED!
{12C95DA2-799D-48AA-802E-9BDFFE19394E}

{E7740416-DDB1-4DD2-8748-98D7B12BF5F2}

@vcdf4497

Yes, I have the same problem, but I don't like GitHub anymore. It used to be better than today, and the Pro version is less advantageous than the Classic version

@immortal71

yeah same goes for me wtf is github doing ,they donot value customer at all ,bring new update resolve this issue or we all r swtiching to other alternatives i can wait 5 hourrs after every 15 min of work duhhh

@Chewpasta

same problem here, 5 prompts and now i have the message "You've hit your session rate limit. Please upgrade your plan or wait a moment for your limit to reset." like dude, are they gonna solve the problem or what

@axonde

And my profile indicates only :

image
You must be logged in to vote

1 reply

@toinemini

+1. I am even on Pro+. I am also rate limited for 3 days :D

FYI, I am on Pro+ and have only hit 14.0% usage FOR THE MONTH so far. Completely unbelieveable! I am not a heavy user at all.

image

this message usually means you hit a short-term rate limit, not your monthly quota. the usage bar on your profile updates with some delay, so it can easily show something like 10–15% while the backend has already throttled you for a burst of requests.

copilot has two separate limits:

the monthly premium allowance (the bar you’re seeing)
and a per-minute / per-hour rate limit to prevent too many requests at once

the error you posted matches the second one. it usually clears on its own after a short cooldown. if it keeps happening even with light usage, it might be the extension sending a lot of background requests without you noticing.

give it a bit of time and it should work again. if it doesn’t, try restarting the editor or checking if any plugin is spamming copilot in the background.

hope this helps! and if it solves it, feel free to mark the thread as answered so others can find it later.
You must be logged in to vote

0 replies

I am also getting it very frequently when assigning multiple issues to coding agent, only coding agent is using the requests. any way to increase the limits like a higher plan. it's really frustrating and negates the purpose of an autonomous agent if i have to come online after every 2 hours and restart the agent. i understand rate-limits are necessary but then at least have a way to schedule auto-retry for coding agents if rate-limited after the cooling period.
You must be logged in to vote

3 replies

@a-man-called-q

I am also getting it very frequently when assigning multiple issues to coding agent, only coding agent is using the requests. any way to increase the limits like a higher plan. it's really frustrating and negates the purpose of an autonomous agent if i have to come online after every 2 hours and restart the agent. i understand rate-limits are necessary but then at least have a way to schedule auto-retry for coding agents if rate-limited after the cooling period.

IMO, rate limiting is to prevent DOS attack I think. I also hit rate limits sometimes. from your words assigning multiple issues to coding agent it seems like the work around is to separate those to multiple agents ? I did that sometimes

@abhisheksiddhu

In the hosted agent, we just assign issues in Github to copilot and it'll start an agent automatically hosted on Github actions. I face the rate limit less frequently if I only assign one issue at a time, but then I have to come online every few hours and assign the next task.

I know that rate limits are needed, that's why I am suggesting to have a feature to queue tasks automatically.

What I want is that I plan an epic with multiple sub-issues with details laid out, assign them all to copilot, come online after a few hours or next day depending on the amount of work, review the generated code and move on. Github should do the queuing starting stopping of agents without my manual intervention.

@nathanielx044-gif

I’m running into the same problem as well. When multiple tasks are assigned to a coding agent, rate limits tend to interrupt the workflow and reduce the overall usefulness of automation.

I understand why limits exist, but it would really help if there were higher thresholds on paid tiers or at least an automatic retry after cooldown. Manually restarting the agent kind of defeats the purpose of an “autonomous” system.

I’ve seen similar ideas discussed in grout guy where processes are handled more smoothly without breaking the flow.

Can more transparency be added so we can keep using your api instead of having to look elsewhere? If you added - you need to stay under x requests per minute or wait 10min/1h etc to continue it would be a much better devx
You must be logged in to vote

4 replies

@novello-dev

@Obazzi

Can we estimate how many per hours we can do before we hit the ratelimit? Would be useful for my Ralph script to put the right amount of sleep between the runs. Actually, could make a ralph script that detects the ratelimits and log them and also progressively make less requests per hour until it doesn't get rate limited anymore....

@advicee12

Do github respond xDD like i bought pro yestoday and after 1 hour and 2-3% requests got rate limited until today xDDDD like today i work for 1 hour and rate limit again xD whats the point to work if you can't work longer than one hour, then you have to wait another 2-5 hours ....

@Florin7222

make less requests per h

Did you do that? would need some help on that cause i am on the 40 dollars plan and funny enough i hit every rate limit, the temp ones, the monthly and recently, the weekly one.. like what.. to use my 1500 requests i need to spend alot of hours on my computer and if i do, i get rate limit daily and weekly. my point being, they make it hard for me to reach my full credit/request with all those limits

I am getting this too and it seems like a you problem. I have 1 instance of Copilot CLI running and doing 1 thing yet I bump into this maybe once or twice a day. It leaves me mid job so I have to dig around and figure out what is now half done or broken and I suspect it is eating a premium request. I don't have any control over how much the AI makes calls but you guys do. You could at least handle this gracefully instead of crashing out mid request. Paying Pro+. I glance over and I see a page of these.

` ✓ Edit ThreeStageFFTConfig.cs

✓ Edit DebugCarryFusedIsolated.cs

✗ Model call failed: {"message":"Sorry, you have exceeded your Copilot token usage. Please review our Terms of
Service.","code":"rate_limited"}
(Request ID: F3E8:12948C:2968905:2FAE35B:694165D9)

✗ Model call failed: {"message":"Sorry, you have exceeded your Copilot token usage. Please review our Terms of
Service.","code":"rate_limited"}
(Request ID: F3E8:12948C:2974A43:2FBB911:69416605)

✗ Model call failed: {"message":"Sorry, you have exceeded your Copilot token usage. Please review our Terms of
Service.","code":"rate_limited"}
(Request ID: F3E8:12948C:297CDED:2FC4B02:69416624)`
You must be logged in to vote

0 replies

Same problem here. I get a lot of rate-limited with ONLY ONE AI working on the project.
i really want to support GitHub/copilot, really...... but the UX is not great, no indication what is going on with AI, usage monitoring and no way to fine-tune your AI usage settings. The throttle before the rate limit would be much better solution but right now it just stops.
This is not great experience and feels a broken system. With this it give a reason to search better options with competitors.
Dont be NOKIA ("We didn't do anything wrong, but somehow, we lost").... fix it before it is too late.
You must be logged in to vote

0 replies

I've run into this and have yet to find the setting that lets me pay for more. Please take my org's $$ and let me continue. I instead end up stalled halfway through the plan. Has anyone tried BYO model to get around this?
You must be logged in to vote

1 reply

@kszpirak

Just run into this exact same wall. Sorry, you have been rate-limited. Please wait a moment before trying again. Learn More

terrible. I can't work like this; same idea; let me pay for more request and freaking take my money let me continue the work.

✗ Model call failed: Sorry, you have exceeded your Copilot token usage. Please review our Terms of
Service. (Request ID:
A572:333BC0:50252FC:68A5913:69560085)

✗ Execution failed: 429 Sorry, you have exceeded your Copilot token usage. Please review our Terms of
Service.

Such a helpful error -_-
I feel like switching to another AI because idk what it's even down for... is this rate limiting or something else?
You must be logged in to vote

0 replies

{ message:
'Sorry, you have exceeded your Copilot token usage. Please review our Terms of Service.',
code: 'rate_limited' } }
deosnt specify how much time should wait, just normal usage no concurrency and i get this error its not normal.
You must be logged in to vote

0 replies

I'm running into this too, and even waiting several minutes, I still get the same rate limiting error with a copilot enterprise account. Your terms of service do not show any rate limit values, just that you will be rate limited if you violate TOS.
You must be logged in to vote

0 replies

They are apparently throttling Pro accounts, making the chat completely unusable. Terrible management and product experience.
You must be logged in to vote

2 replies

@Wulfic

yeah and it eats up your overall usage, so you can get halfway through a chat session and it just stop midwway withh your code butchered and then they count it against you anyway. im getting ready to switch directly to anthropic or something. its a total waste of time and energy with the new changes they made recently

@umezakip

Just another reason for me to cancel my subscription. Theyre getting to greedy.

Model call failed: Sorry, you have exceeded your Copilot token
usage. Please review our [Terms of Service](https://docs.github.co
m/en/site-policy/github-terms/github-terms-of-service). (Request
ID: 90E2:1C68C2:F8DEB:119647:69667B7C)

Meanwhile I've got pro, and added 500$ budget with my creditcard because I use this a lot
But yet I'm no where near my budget and still getting this message
You must be logged in to vote

0 replies

Yeah, this is nonsense. I'm here cause I'm currently wasting time waiting for the throttling to reset. I gave it one task: to refactor a class that had grown into a god class. I've hit the rate limit twice on the same task now. Maybe it should look at the request scope and at least say "Sorry Dave, I can't do that" rather than the unfriendly/useless way it's handling this issue. I've got a pro plan and there's a lot of credit left on it, so this is unacceptable. Do better.
You must be logged in to vote

1 reply

@alexmatjuhin

I'm here because of the opposit problem: around 400 Queries and Commands with duplicated logic which I moved into the middleware in my pipeline so the only thing I've asked is to remove those duplicates and refactor the rest a bit to follow common principles and the latest C# syntax sugar. And boom! The chat hangs with 429 error In the middle of refactoring pack of next 10 files. Monthy usage is 47%. Every metric here is in "included" zone. Those hourly and daily limits should be more clear!

I have also run into this issue multiple times before. Usually, this happens after excessive usage of one of the more premium models. A good way to tell this is to look at the right-hand side of the model selection UI:
image
The number that you see on the right indicates the current amount of rate usage a model consumes, and a more premium model (ex. the Claude 4.5 Opus I am using) will have a higher number, and thus hit the rate limit faster. Do not worry, because they usually reset within 24 hours, so you should gain access soon.
You must be logged in to vote

2 replies

@KlomDark

If I was driving my car on the interstate and it randomly, with no warning, just pulled to the side of the road and made me wait 24 hours to drive again, I would certainly not find that acceptable. Luckily it seems the reset is usually at the top of each hour rather than 24 hours. But this is still unacceptable for a paid plan. If it was free I'd understand, but it's not free.

@elihuvillaraus

Can not agree more! Those who say "Just wait" do not understand that this is work, money, pro stuff, not hobby nor games.

Bait and switch scam

You've used 69% of your weekly rate limit. Your weekly rate limit will reset on April 26 at 8:00 PM. Learn More
You must be logged in to vote

0 replies

I'm having the same alert. So, I think is the last thing I needed to go for Claude. Good bye copilot.
You must be logged in to vote

0 replies

SHIT WEEKELY LIMIT! I have requests, but can not use it.
You must be logged in to vote

1 reply

@MouradiSalah

You've used over 90% of your session usage limit. Your limit resets in 7 minutes
You must be logged in to vote

0 replies

same issue with less than 1 hour
You must be logged in to vote

0 replies

same issue just now :
Hit weekly rate limit last week
Waited for the weekly reset
Now immediately hitting session rate limit
You must be logged in to vote

0 replies

I still have 200 premium model calls remaining in my quota. Today is April 28th, yet I have been blocked from using premium models for five consecutive days. This means I will not be able to use my remaining premium quota at all.
I only started using this subscription on April 19th, but you plan to reset my quota on May 1st and clear my nearly 200 unused premium credits. This is extremely unfair. My subscription plan is valid until May 18th, and you have no right to reset users’ benefits based on a calendar month cycle.
I now have two clear demands:
If my access to premium models cannot be restored today, you must either process a refund or reset my full call quota immediately.
You have no right to deprive paying users of their entitled plan benefits. I would prefer a refund first. At the very least, I still have around 6 USD of unused service value.
The current unreasonable rate limit policy has severely damaged the user experience and made me extremely dissatisfied. A five-day usage ban is entirely unreasonable and inappropriate. @github @github copilot
You must be logged in to vote

1 reply

@jadmadi

Somewhat similar situation, opened a ticket, and then decided to cancel. It's a hopeless case basically.

You've used 69% of your session rate limit. Your session rate limit will reset on April 30 at 11:24 AM. Learn More
You must be logged in to vote

0 replies

I paid for the service but had to wait 5 days because of the limit. I only used a few requests before reaching the limit and had to wait 5 days. This is a terrible service.
You must be logged in to vote

0 replies

just happened to me . ive used 0% of my limit. wtf
You must be logged in to vote

0 replies

Sorry, you have been rate-limited. Please wait a moment before trying again. Learn More

Server Error: Sorry, you've exceeded your 5 hour session limits. Please review our Terms of Service. Error Code: user_global_rate_limited:pro

SERIOUSLY GITHUB, I HOPE YOUR TEAM GETS HIT WITH A CLASS ACTION LAWSUIT FOR DISRUPTING SERVICE !

YALL RUINED COPILOT FOR PEOPLE WHO USED TO VIBE CODE!
You must be logged in to vote

0 replies

Here's what you do'
Go to platform.deepseek.com, sign up for an api key, top it off with $10 of credits.

Install this extension into vscode'

https://marketplace.visualstudio.com/items?itemName=Vizards.deepseek-v4-for-copilot

Enter your deepseek api key.
Select deepseek v4 flash

Keep working in vscode copilot for $0.032 per million input tokens.

Marvel at how much you can get done for $10 with no interruptions or rate limits.

Enjoy!
You must be logged in to vote

1 reply

@tommyziegler2003

Great idea. Tho it's also free if you can host it via ollama. I might try it.

If you still use github copilot i suggest to use a same country vpn when you meet the rate limit. You can skip the stupid limiters that way. I refunded 8mc of sub at the tome of this comment i use ollama and im happy :)
You must be logged in to vote

0 replies

You must be logged in to vote

0 replies

Im not able to use all of the 100% of my TOKENS becuase allways when I try to work im rate lmited !!
"Error: (rate_limit) You've reached your weekly rate limit. Please wait for your limit to reset on May 18, 2026 at 1:59 AM or switch to auto model to continue. Learn More (https://docs.github.com/en/copilot/concepts/rate-limits). (Request ID: E1CE:247937:2AE10C0:3114724:6A08110C)"

WHATS THIS ?!!!
You must be logged in to vote

2 replies

@xAstroBoy

Github being greedy jackasses , that's what is it, ruined a useful tool for everybody!

@SagaciousFish

It's OK. With the new cost system, you'll be using up all of the 100% of your TOKENS in no time. 😁