Claude with Copilot Chat is broken

[unitythemaker]

Select Topic Area

Bug

Body

Claude 3.5 Sonnet model used from GitHub Copilot Chat is very broken on both VSCode and Zed Editor. It seems to almost always cut its messages in the half while streaming. It's very unreliable compared to GPT-4o. GPT-4o has more frequent "ticks", GPT-4o keeps streaming more often almost for each token whilst on Claude, it's more like every 15-20 tokens at a time in longer periods. Problem is, I almost always prefer Claude over GPT-4o for its helpfulness, giving actual work instead of giving examples for me so I don't want to use GPT-4o and I need this issue to be fixed.

In my opinion, the source of the problem is not client-sided but it's server-sided. I think there is some bug or misimplementation between Claude 3.5 Sonnet API and Copilot servers. Because, when I am using the Anthropic API directly from the Zed Editor, it works perfectly and similar to GPT-4o, more frequent "ticks".

The issue arises more often as the context size is larger (e.g. 20k).

[github-actions[bot]]

💬 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. ⭐

[GitMurf]

Claude 3.5 Sonnet model used from GitHub Copilot Chat is very broken on both VSCode and Zed Editor. It seems to almost always cut its messages in the half while streaming. It's very unreliable compared to GPT-4o...
In my opinion, the source of the problem is not client-sided but it's server-sided. I think there is some bug or misimplementation between Claude 3.5 Sonnet API and Copilot servers. Because, when I am using the Anthropic API directly from the Zed Editor, it works perfectly and similar to GPT-4o, more frequent "ticks".

Completely agree. I am having the exact same problems with claude through copilot chat, and several people that I know are also having the same.

[robertodauria]

I am experiencing identical issues with Copilot + Claude 3.5. The responses are truncated after a few lines, even when sending tiny contexts. I am sure that I am not exceeding any acceptable usage limits, as I have not used Copilot since December, and I am sending just a few 400-token requests. This same request works without any issue when sent to the Zed backend (Anthropic API).

[bassosimone]

I also experience similar issues as the ones described above with Copilot + Claude 3.5.