GitHub Community
Copilot update: rate limits + fixes #190176
Replies: 23 comments 12 replies
-
|
What would be helpful is if there were some sort of auto restart once the limit was removed or a long running process flag that would slow things down on the client based on current overall load on the server. I don't mind the rate limiting so much as the disruption it causes. One frustration I had this morning... I run each story (I use the BMAD Method framework) in a subAgent in a separate worktree. When I resumed, it decided to continue its work in my main branch outside of a subAgent 😠 |
Beta Was this translation helpful? Give feedback.
-
|
The recent changes in GitHub Student Copilot makes absolutely zero sense. You give us access to the actual capable models (Claude Opus, Sonnet) through the Student Pack. We integrate it into our workflows, we start relying on it to learn and build proper architecture... and then overnight, you silently restrict access to push everyone toward the paid premium tiers. If the goal is just to use students for an upsell funnel, you might as well completely kill your useless, nerfed Student Pack. Giving us a crippled tool that throws model_not_supported errors the second we need to do real work is just insulting. Typical Microsoft move. Give with one hand, take with the other. Do better. |
Beta Was this translation helpful? Give feedback.
-
|
Thanks for the transparency and detailed breakdown — really appreciated. That said, I think the core frustration from users wasn’t just about the limits themselves, but how suddenly and broadly they were enforced. Since the fix applied system-level limits across all models, it disrupted even normal workflows, which made it feel unpredictable from a user standpoint. A couple of thoughts that might improve the experience going forward: Graceful degradation instead of hard blocking Overall, it’s good to see quick mitigation and acknowledgment. Hopefully, future changes can balance system protection with a smoother user experience. |
Beta Was this translation helpful? Give feedback.
-
|
Still experiencing rate-limits when using 5.4 and building apps. It occurs after some time of frequent use. |
Beta Was this translation helpful? Give feedback.
-
|
Switching agent models doesn't fully work: I just switched from Claude to GPT 5.3 for the agent run, and the agent now works, but the automatic review it starts as the last step is still rate limited and runs into 429. The agent also says it cannot control the LLM of the code-review tool. So no agent can complete its work properly. In general, the code-review rate limit should be much more transparent, it confuses the agents trying to use it: |
Beta Was this translation helpful? Give feedback.
-
|
I'm still experiencing rate limit errors as well. I am not doing anything crazy, and only using one agent at a time, no sub-agents. Please clarify if receiving these errors still counts against the usage. Requests with errors like this should be refunded. |
Beta Was this translation helpful? Give feedback.
-
Oddly, today seems to be better for me 🤷♀️ |
Beta Was this translation helpful? Give feedback.
-
|
Beta Was this translation helpful? Give feedback.
-
|
These rate limit changes are too drastic. I'm finding it nearly impossible to work efficiently anymore. |
Beta Was this translation helpful? Give feedback.
-
|
I'm still experiencing rate limit errors as well. I am not doing anything crazy, and only using one agent at a time, no sub-agents. Please clarify if receiving these errors still counts against the usage. Requests with errors like this should be refunded. These rate limit changes are too drastic. I'm finding it nearly impossible to work efficiently anymore. |
Beta Was this translation helpful? Give feedback.
-
|
Agreed |
Beta Was this translation helpful? Give feedback.
-
|
Further frustrations... I'm running GitHub agents and subAgents. The agent is mostly a driver to spawn the subAgents that run sequentially (i.e., I'm not attempting parallelism, or I'd have even more issues) and run within my own defined worktree. One worktree per story. I get rate-limited in the middle of a subAgent, which shows no indication that I've been rate-limited other than that everything stops. I can't tell if that's because the AI process got confused (lost context) about where it is, or because it is rate-limited, now, it is more often because of the rate limit. Some time later, I type continue, and it picks up where it left off, but it has forgotten it is in a worktree and starts processing (or should I say reprocessing) the story in the main branch/worktree, and therefore starts the story over from scratch. Now that I know this, I just start my process over from the top and let it know what story it is on. But you'd think "continue" would pick up where it left off. As I said above, I don't begrudge the rate-limiting but it would be great if the client side knew how to work with it better. Also, I'd be happy to work with you all to make this work smoother. |
Beta Was this translation helpful? Give feedback.
-
|
Any idea after how much time it start working when the user is shown with rate limit error. |
Beta Was this translation helpful? Give feedback.
-
|
This update has been a very poor experience. Claude is overthinking far too much, and I’m completely not used to it. The frequency control is disrupting my work and making it impossible to proceed normally. |
Beta Was this translation helpful? Give feedback.
-
|
SOLUCIONEN ESTO |
Beta Was this translation helpful? Give feedback.
-
|
Did you resolve it? |
Beta Was this translation helpful? Give feedback.
-
|
Copilot has done barely any work this morning and it is already striking... 🥐 |
Beta Was this translation helpful? Give feedback.
-
|
+1, I switched to pro+ only to get this |
Beta Was this translation helpful? Give feedback.
-
|
Yes I am on pro+ too :) |
Beta Was this translation helpful? Give feedback.
-
|
Did you resolve it? |
Beta Was this translation helpful? Give feedback.
-
|
no it is obviously a bug on copilot side |
Beta Was this translation helpful? Give feedback.
-
|
back working for me |
Beta Was this translation helpful? Give feedback.
-
|
I made my 1st request to add one single unit test and got rate limited, switching models doesn't fix anything, apparently the bug is still not fixed. I'm seriously considering unsubscribing if this isn't fixed ASAP. |
Beta Was this translation helpful? Give feedback.
-
早上第一次请求就这样了！！怎么处理呢
|
Beta Was this translation helpful? Give feedback.
-
|
每次重试也是会扣用量的！pro+ |
Beta Was this translation helpful? Give feedback.
-
|
你们解决了吗？ |
Beta Was this translation helpful? Give feedback.
-
I have used difference models but it has same problem. You are wrong |
Beta Was this translation helpful? Give feedback.
-
|
Are there any other better alternatives? |
Beta Was this translation helpful? Give feedback.
-
|
I'm a GitHub Copilot Pro+ user I just woke up and started my workday, and I’ve already triggered the usage limit. I only sent two extremely small requests:
That’s it. Both were minor tasks. |
Beta Was this translation helpful? Give feedback.
-
why？
|
Beta Was this translation helpful? Give feedback.
-
|
5 minutes into session and i got ratelimited? how? |
Beta Was this translation helpful? Give feedback.
-
|
codex: hello |
Beta Was this translation helpful? Give feedback.
-
|
It interrupted my session. Is it possible to control the rhythm smoothly instead of abruptly interrupting the current session? |
Beta Was this translation helpful? Give feedback.
-
|
Imagine rejecting requests from paying customers... |
Beta Was this translation helpful? Give feedback.
-
|
I'm equally frustrated, like everyone else, even when paying for extra tokens in my budget. One would think that blocking usage would prevent income from those willing to pay extra! I am confused. |
Beta Was this translation helpful? Give feedback.
and others
Uh oh!
There was an error while loading. Please reload this page.
-
Hey folks, given the large increase in Copilot users impacted by rate limits over the past several days, we wanted to provide a clear update on what happened and to acknowledge the impact and frustration this caused for many of you.
What happened
On Monday, March 16, we discovered a bug in our rate-limiting that had been undercounting tokens from newer models like Opus 4.6 and GPT-5.4. Fixing the bug restored limits to previously configured values, but due to the increased token usage intensity of these newer models, the fix mistakenly impacted many users with normal and expected usage patterns. On top of that, because these specific limits are designed for system protection, they blocked usage across all models and prevented users from continuing their work. We know this experience was extremely frustrating, and it does not reflect the Copilot experience we want to deliver.
Immediate mitigation
We increased these limits Wednesday evening PT and again Thursday morning PT for Pro+/Copilot Business/Copilot Enterprise, and Thursday afternoon PT for Pro. Our telemetry shows that limiting has returned to previous levels.
Looking forward
We’ll continue to monitor and adjust limits to minimize disruption while still protecting the integrity of our service. We want to ensure rate limits rarely impact normal users and their workflows. That said, growth and capacity are pushing us to introduce mechanisms to control demand for specific models and model families as we operate Copilot at scale across a large user-base. We’ve also started rolling out limits for specific models, with higher-tiered SKUs getting access to higher limits. When users hit these limits, they can switch to another model, use Auto (which isn't subject to these model limits), wait until the temporary limit window ends, or upgrade their plan.
We're also investing in UI improvements that give users clearer visibility into their usage as they approach these limits, so they aren't caught off guard.
We appreciate your patience and feedback this week. We’ve learned a lot and are committed to continuously making Copilot a better experience.
Beta Was this translation helpful? Give feedback.
All reactions