A routing switchboard sends coding tasks into different model lanes from an admin control plate
A routing switchboard sends coding tasks into different model lanes from an admin control plate
+ Microsoft News

Copilot auto model selection turns routing into an admin setting

GitHub Enterprise admins can now default Copilot conversations to auto model selection through managed-settings.json, tying model routing to enterprise AI standards.

GitHub Enterprise administrators can now set model to auto in enterprise managed-settings.json so new Copilot conversations start with auto model selection.

The setting is part of a broader managed-settings system that GitHub also made generally available on July 1. Enterprise Cloud customers can keep managed-settings.json in a selected organization’s .github-private repository. GitHub says the configuration applies in VS Code and Copilot CLI for users licensed through Copilot Business or Copilot Enterprise.

This moves model routing from a developer habit into an enterprise standard.

Auto routing becomes a default, not just a user choice

GitHub says administrators can add auto to .github-private/.github/copilot/managed-settings.json so new conversations begin with Copilot auto model selection. Users can still switch to a different model for a specific conversation.

That balance is important. The enterprise can make auto routing the starting point without fully removing local discretion. The admin goal is consistency: reduce the number of users picking models by habit, hype, or stale preference when the system can route based on task.

The managed-settings post gives the control plane. Supported keys include extraKnownMarketplaces, enabledPlugins, strictKnownMarketplaces, disableBypassPermissionsMode, and model. GitHub says supported file-based user configuration is overridden for those keys, and the managed configuration is fetched when a user authenticates, stored in memory, and refreshed hourly.

Routing and governance are merging

This update fits GitHub’s larger Copilot control pattern. The same week included session streaming, credit controls, browser tools, vision, and Kimi K2.7 Code as a new selectable model. Capability is expanding, and so is the administrative layer around it.

For engineering leaders, model routing is no longer only a quality question. It is a policy question. Which model is allowed for which task? What happens by default? Who can override? How do those choices show up in audit, billing, and incident review?

Auto selection can be useful because developers do not always know which model is best for a task. It can also hide a decision that teams may want to understand. A routed answer still needs governance around allowed providers, data handling, cost, and task type.

Sources

The AI Feed Desk

The AI Feed Desk

Editorial desk

The AI Feed Desk tracks AI provider updates, model releases, agent tooling, and enterprise adoption, turning fast-moving announcements into source-linked context for builders and operators.

Noticed a typo, incorrect information, or translation error?

Tell us so we can fix it.

Help Improve This Article

Related Articles

Shared AI credit tokens flow through team lanes while one cost center lane is stopped by a cap gauge

GitHub cost centers can cap shared AI credit pools

GitHub added AI credit pools for cost centers, letting enterprise admins cap how much monthly included Copilot credit one group can draw before metered budgets apply.

The AI Feed Desk

By The AI Feed Desk

Developer tasks pass through a policy gate into several abstract model routing lanes with a cost meter below

GitHub turns Copilot model choice into a governed routing layer

GitHub's July 1 Copilot changes put model routing, browser tools, vision, Kimi K2.7 Code access, and AI-credit caps under clearer admin control.

The AI Feed Desk

By The AI Feed Desk

An abstract coding agent session timeline flows into a secure audit collector

GitHub exposes Copilot agent sessions for enterprise audit

GitHub's public preview lets enterprise owners stream or retrieve Copilot agent session data, including prompts, responses, and tool calls, across major Copilot clients.

The AI Feed Desk

By The AI Feed Desk

A usage dashboard shows AI credit consumption bars for individual developers

GitHub adds per-user AI credit metrics for Copilot admins

GitHub's Copilot usage metrics API now reports per-user AI credit consumption, giving enterprise and organization admins a clearer signal for adoption and budget planning.

The AI Feed Desk

By The AI Feed Desk

A pull request review panel connects to Jira cards and a guarded plugin marketplace

GitHub turns Copilot governance into a code-review and Jira story

GitHub's June 25 Copilot updates pair cheaper code-review analysis with Jira agent visibility and stricter controls over plugin marketplaces.

The AI Feed Desk

By The AI Feed Desk