Fusion repeatedly selects Kimi K3 as the only option in the available model pool and has consumed 75% of my monthly quota in one day (so far)

[Caramel-Andrew]

🏷️ Discussion Type

Bug

💬 Feature/Topic Area

Copilot in GitHub

Body

Description:

Using Fusion mode in GitHub Copilot through VS Code, Fusion repeatedly reports:

"Copilot Fusion selected Kimi K3 (GitHub) because it was the only option based on the available model pool."

However, my model selector shows 13 available models (14 including one retired model), and there appears to be no obvious reason that only Kimi K3 would qualify for the type of work being performed.

During use, I observed the agent repeatedly reading the same file multiple times until I intervened and gave it a gentle slap. My monthly usage allowance increased from 0% to approximately 75% within a single day of normal development work.

I used Copilot heavily during the previous billing cycle and only exhausted my quota two days before renewal. During the current billing cycle, approximately 75% of my monthly allowance was consumed on the first day despite performing a similar volume and type of development work.

It is unclear whether the high usage is related to Fusion's model selection, repeated file analysis, or another issue, but the behaviour differs significantly from the previous month.

Steps to reproduce:

1. Open a project in VS Code with GitHub Copilot enabled.
2. Open GitHub Copilot Chat and select Fusion mode.
3. Use Agent mode to analyse the codebase and perform normal development tasks.
4. Observe Fusion's model selection messages during task execution.
5. During multiple interactions, Fusion reported: "Copilot Fusion selected Kimi K3 (GitHub) because it was the only option based on the available model pool."
6. Compare this message with the models visible in the model selector (13 available models in my case).
7. Monitor agent behaviour and monthly quota consumption during normal use.

Additional observations:

The model selector shows 13 available models (14 including one retired model).
The agent was observed repeatedly reading the same file multiple times during some tasks.
Monthly usage increased from 0% to approximately 75% within a single day.
Similar development work during the previous billing cycle consumed quota at a significantly lower rate.

Expected behavior:

• Fusion should select from the available model pool in a way that is consistent with the models available to the user, or clearly explain why only a subset of models are eligible.
• Fusion should avoid repeatedly analysing the same unchanged file unless additional analysis is required.
• Monthly quota consumption should be broadly proportional to usage and should not differ dramatically from previous billing cycles when performing a similar volume and type of development work.
• The reason for model selection should be sufficiently transparent that users can understand why a particular model was chosen.

Version Reference

VSCode
Version: 1.145.0 (user setup)
Commit: a717861c07586a53d83b80d742cd3f59b20a6273
Date: 2026-11-03T15:03:51-08:00
Electron: 43.8.3
ElectronBuildId: 15784459
Chromium: 150.0.7871.250
Node.js: 24.22.1
V8: 15.0.245.31-electron.0
@github/copilot: 1.0.103.r41379698026.g3219c66
@github/copilot-sdk: 1.0.27.41379698026.g3219c66
OS: Windows_NT x64 10.0.26300

Github Copilot
Account Level: Pro
Github Copilot Chat Extension: 0.73.0

NOTE This discussion was accidentally posted using my company GitHub account, which does not have a Copilot subscription associated with it. The behaviour described in this report was observed while using my personal GitHub Copilot Pro subscription.

[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. ⭐

[jpvignolo]

I'm experiencing the exact same issue since Fusion replaced Auto. For me it seems to happen only with fusion in standard mode. With High it seems to select simpler models for simpler tasks. And yes it's burning a lot more credits for nothing compared to Auto.