Pro+ user hitting “weekly rate limit” after ~300/1500 requests — system unusable #192843
Replies: 1 comment
-
|
💬 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 ⏩
Where to look to see what's shipping 👀
What you can do in the meantime 💻
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. ⭐ |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
🏷️ Discussion Type
Bug
💬 Feature/Topic Area
Copilot in GitHub
Body
This is becoming unusable for real work.
I’m a Pro+ user and have only used around 300 out of 1500 requests, yet I’m constantly getting the error:
“you've exceeded your weekly rate limit.”
In practice, I can send 1–2 messages, then I’m blocked again for hours. After waiting, I can again send 1–2 messages, and the same issue repeats. This behavior makes no sense compared to the visible quota.
There are clearly multiple hidden limits (weekly, burst, model, etc.), but users have zero visibility into any of them. As a paying user, I should at least know:
which limit I actually hit
why it was triggered
when it will reset
Right now it feels like a complete black box. The system behaves randomly — sometimes it works, sometimes it just stops with no explanation.
I also tried switching to the “Auto” model, but that introduced another serious issue. When it works, the responses are often inaccurate or completely off from what I actually asked. In many cases, it generates irrelevant output or makes obvious mistakes that require constant correction.
So I’m effectively stuck between:
models that are rate limited after 1–2 messages, and
a fallback model that is unreliable and produces poor-quality results
Rate limiting itself is understandable, but this level of lack of transparency and unpredictability makes Copilot unreliable for professional use.
Beta Was this translation helpful? Give feedback.
All reactions