GitHub Copilot coding agent, our asynchronous developer agent, now works with Claude Sonnet 4.5, Anthropic’s most advanced model for coding and real-world agents.

Claude Sonnet 4.5 is state of the art on the SWE-bench Verified evaluation, which measures real-world coding abilities.

If you are a Copilot Pro or Copilot Pro+ user, coding agent will automatically use Claude Sonnet 4.5.

If you are a Copilot Business or Copilot Enterprise user, an administrator will need to enable the Claude Sonnet 4.5 policy in Copilot settings. If this policy is not enabled, coding agent will continue to use Claude Sonnet 4.

To learn more about Copilot coding agent, head to our documentation on coding agent.