Copilot Cloud Agent resolves GPT-6 Luna, then fails on unavailable GPT-5.4 mini #211120
Replies: 2 comments
|
💬 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. ⭐ |
|
This behavior points to a model mismatch bug during the initialization lifecycle of the Copilot Cloud Agent, where the configuration layer correctly resolves the target model, but the internal validation/orchestration layer falls back to a different default. Look for Agent Manifest and check Run History. |
Uh oh!
There was an error while loading. Please reload this page.
🏷️ Discussion Type
Bug
💬 Feature/Topic Area
Copilot Cloud Agent
Body
Summary
A GitHub Copilot cloud-agent run failed before the task started. The selected model resolved to GPT-6 Luna, but runtime validation then attempted to use GPT-5.4 mini and stopped with an unavailable-model error.
Evidence
env="sweagent-capi:gpt-6-luna" settings="sweagent-capi:gpt-6-luna", resolving togpt-6-luna.Model "gpt-5.4-mini" is not available.Expected
The model selected and resolved for the cloud-agent run should be used consistently through runtime setup, and the agent task should start.
Could the Cloud Agent runtime be retaining a stale GPT-5.4 mini fallback or validation reference?
All reactions