GitHub Copilot Retires Claude Opus 4.8, Opus 5 and GPT-5.6 Terra

GitHub removed Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra from most Copilot experiences on 16 November 2026. Suggested replacements are Claude Opus 5.5 and GPT-6.1 Sol or GPT-6 Luna, enabled by default for Business and Enterprise unless admins opted out.

Christopher Ort
By Christopher Ort · 2 min read
Share
A developer's desk at night seen over the shoulder.
Image: AI-generated, i10X
On this page
  1. Quick Take
  2. In plain English
  3. What happened
  4. What to do
  5. Why it matters
  6. What's still open
  7. Sources
  8. i10x Perspective

GitHub has removed three older AI models from most of Copilot. As of 16 November 2026, Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra are deprecated across Copilot Chat, inline edits, ask and agent modes, and code completions. GitHub points users to Claude Opus 5.5 and to GPT-6.1 Sol or GPT-6 Luna instead.

i10X Quick Take

Quick Take

  1. Removed: Claude Opus 4.8, Claude Opus 5 and GPT-5.6 Terra, from most Copilot experiences.
  2. Replacements: Claude Opus 5.5 for both Opus models; GPT-6.1 Sol or GPT-6 Luna for GPT-5.6 Terra.
  3. Exception: Claude Opus 4.8 stays available to individual Copilot subscribers on annual plans.
  4. Fallback: Claude Opus 5 is gone from the model picker but remains available for runtime refusal fallback through existing authorized access.
  5. Admins: the suggested alternatives are enabled automatically for Copilot Business and Enterprise unless an administrator turned off the global default or disabled the model.

In plain English

GitHub Copilot lets developers choose which AI model writes and reviews their code. GitHub has now taken three older models off that menu and is steering people to newer ones from Anthropic and OpenAI. If your team picked one of the old models by name in a workflow or integration, you should switch it to a supported model.

What happened

GitHub announced the deprecations on 15 October 2026 and carried them out on schedule. “Please update your workflows and integrations to use supported models,” the company wrote in its changelog. “No action is required to remove the deprecated models.”

Existing enterprise model policies continue to apply. If an organization turned off the global default for new models, admins must enable the replacements manually through model policies in Copilot settings. Once enabled, the models show up in the Copilot Chat model selector.

What to do

Developers: check any saved prompts, custom agents or scripts that request a model by name. Requests for removed models will no longer work in most experiences.

Admins on Business and Enterprise plans: confirm whether the global default is on. If not, enable Claude Opus 5.5, GPT-6.1 Sol or GPT-6 Luna for the teams that need them.

Individuals on annual plans who rely on Claude Opus 4.8 can keep using it for now.

Why it matters

Model churn is now a normal part of running coding agents. Teams that hard-code a model into CI pipelines or agent configs, for example with tools like JetBrains’ TeamCity plugin for Codex and Claude Code, need a process for swapping models when vendors retire them.

The replacements also say something about where developers are going. Claude Opus 5.5 is already the default pick for many coding tools, as our look at Claude Code vs Codex usage in T3 Code showed.

What’s still open

  • How long Claude Opus 4.8 will remain for annual-plan individuals.
  • Whether per-request costs differ between the removed models and their replacements under Copilot’s premium request system.

Sources

  • GitHub Changelog, “GitHub Copilot models deprecated on November 16, 2026,” 16 November 2026
  • GitHub Changelog, “Upcoming deprecation of selected GitHub Copilot models in mid-November,” 15 October 2026

i10x Perspective

A small changelog entry, but a reminder that the model under your coding assistant is a moving part. Teams that treat model choice as configuration, not code, will handle these retirements in minutes. Everyone else finds out when a pipeline breaks. More in our Coding section.

Christopher Ort

Christopher Ort

Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU). He is the author of “The Style Penalty”, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.

CodingNews

Editor's picks

  1. Microsoft Teams Lets AI Agents Join as Digital Teammates With Their Own Identity
  2. Azure Cobalt 200 VMs Reach General Availability With Up to 50% More CPU Performance
  3. Microsoft Launches AI Transparency Initiative With NIST Standards Work and 40 Partners
  4. Agent 365 Gets Business Analytics and FinOps for AI as Nadella Cites 50 Million Agents
  5. Microsoft Perception Is Generally Available, With Red, Blue and Green Security Agents