TL;DR: On Oct 19, 2026, GitHub deprecated six models across most GitHub Copilot experiences: Gemini 3.7 Flash, GPT-5.5, GPT-5.4, GPT-5.4 mini, GPT-5 mini and Grok 4.5. The suggested replacements are now newer than the ones GitHub named when it announced the change in September, so check which models your team actually has turned on. On the same day, GitHub made a REST API for Copilot model policies generally available, so admins can check and change model access in bulk.

What changed

GitHub’s Oct 19 changelog says the models are deprecated “across most GitHub Copilot experiences (including Copilot Chat, inline edits, ask and agent modes, and code completions).” The deprecations were first announced on Sept 18, 2026. A second round is already scheduled: GitHub’s Oct 15 notice says Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra will be deprecated on Nov 16, 2026, though Claude Opus 4.8 stays available to individual Copilot subscribers on annual plans.

Deprecated model Suggested alternative (Oct 19) Suggested alternative (Sept 18 notice)
Gemini 3.7 Flash Gemini 3.8 Flash Gemini 3.8 Flash
GPT-5.5 GPT-6.1 Sol GPT-5.6 Sol
GPT-5.4 GPT-6.1 Sol GPT-5.6 Sol
GPT-5.4 mini GPT-6 Luna GPT-5.6 Luna
GPT-5 mini GPT-6 Luna GPT-5.6 Luna
Grok 4.5 Grok 4.7 Grok 4.6

GitHub says no action is needed to remove the old models. The catch is the replacements. In September, GitHub said the suggested alternatives would be turned on automatically for Copilot Business and Enterprise customers under default model enablement, unless an admin had turned off the global default or disabled the model. The Oct 19 post names newer models and says Copilot Enterprise admins “may need to enable access to alternative models through their model policies in Copilot settings.”

The new model policy API

Also on Oct 19, GitHub made GET and PATCH endpoints for Copilot model policies generally available at four levels: enterprise, enterprise teams, enterprise apps and organization. For example, an organization’s policies live at /orgs/{org}/copilot/model-policies. That means you can check which models are enabled across many organizations with a script instead of clicking through settings.

What to do

  1. Open the model picker. In VS Code or on github.com, check that the suggested replacement for each model your team used appears in the Copilot Chat model selector.
  2. If it’s missing, enable it. Admins can turn on the replacement in the model’s policy in Copilot settings, or with the new model policy API.
  3. Update anything that names a model. Scripts, agent configs, team docs and prompts that pick GPT-5.5, GPT-5.4, GPT-5 mini, Gemini 3.7 Flash or Grok 4.5 should move to a supported model.
  4. Re-check costs and limits. A newer model can behave and bill differently. If your team uses Copilot with other coding tools, our guide to Claude and Codex usage limits covers how those plans meter usage.

Sources & method

Sources

Method. Last checked October 20, 2026. Model names, dates, alternatives and API endpoints come from GitHub’s changelog posts. No affiliate links on this page. How we work.

Spotted an error? See our corrections policy and tell us.