Decline in Copilot functionality (Certain Models) #191422
Replies: 4 comments 1 reply
-
|
💬 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.
-
|
+1. I’m seeing the exact same "stale agent" issue with Opus failing to complete tasks. Yesterday, it was constantly throwing timeouts. Today, it seems like a band-aid was applied so it just finalizes the task at the last second with zero progress. Honestly, it feels deliberate—like a nasty workaround to force us to burn through our Premium Requests on a broken service without throwing an explicit error. We shouldn't be losing our quota on this. |
Beta Was this translation helpful? Give feedback.
-
|
Worse still, while the throttling used to cause flat-out timeouts, GitHub's alleged 'fix' is a joke. Now, the system just stalls and manages to spit out a completion right before the timeout limit hits—barely making any real progress. Meanwhile, Sonnet runs circles around it, doing way more in less time. This janky workaround is probably because Copilot is struggling with per-token costs, but to the user, it just exposes how unreliable the tool is and how GitHub is failing to deliver. |
Beta Was this translation helpful? Give feedback.
-
|
Hey! The issues you're describing (over-analyzing, stalling, ignoring prompts) actually align with documented infrastructure incidents on Anthropic's end from late March — there were elevated errors and timeouts on Opus 4.6 from March 26–31 that could cause exactly this behavior. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
🏷️ Discussion Type
Product Feedback
💬 Feature/Topic Area
Copilot Agent Mode
Body
Whatever has been done in the past week has lobotomized Claude Opus 4.6. It burns credits and spends time over analyzing with no actual progress. It often gets stuck and stalls the whole process. I often have issues where i type a prompt that it completely disregards and decides to work on something from the previous prompt instead. Its wild how all these errors are happening more and more as my subscription runs on yet there is no refund for burned credits wasted on something that clearly isnt functioning as intended.
Beta Was this translation helpful? Give feedback.
All reactions