The Copilot model deprecation announced on September 10, 2026 removes the MAI-Code-1-Flash model from all Copilot experiences and directs users to the newer MAI-Code-1.1-Flash model. Engineers and operators need to adjust any references, policies, or automation that still target the retired model to avoid unexpected fallback behavior.
What changed
GitHub has officially retired the MAI-Code-1-Flash model across Copilot Chat, inline edits, ask and agent modes, and code completions. The suggested replacement is MAI-Code-1.1-Flash. No automatic removal is performed; the old model simply becomes unavailable for new requests.
Impact on engineering workflows
Any CI/CD pipelines, custom scripts, or integration points that explicitly request MAI-Code-1-Flash will now receive an error or fallback to a default model. Updating those references to MAI-Code-1.1-Flash restores expected behavior. The change also affects any internal documentation or onboarding material that lists the deprecated model as an option.
Operational steps for administrators
- Review model policies in the Copilot Enterprise settings to ensure
MAI-Code-1.1-Flashis permitted. - Enable the new model in the policy if it is not already allowed.
- Confirm visibility by opening the model selector in VS Code or on github.com; the new model should appear once the policy is active.
- Communicate the change to teams that rely on Copilot‑generated code, advising them to update any hard‑coded model identifiers.
- Enterprise customers with questions can contact their account manager for assistance.
Related CloudNinjas coverage: AI engineering.
What This Means For Practitioners
Treat the deprecation as a prompt to audit all Copilot model references in your codebase and automation. Enable MAI-Code-1.1-Flash through your model policies, verify its availability in the UI, and test critical workflows to confirm the new model meets functional expectations. Keep an eye on future GitHub model announcements to avoid similar disruptions.
