Grok 4.6 Is Now Available in GitHub Copilot
xAI’s Grok 4.6 has begun rolling out inside GitHub Copilot, GitHub says, positioning it as a reasoning model built for agentic coding and complex, multi-step workflows. This changelog entry is notable beyond the feature itself: it’s the first confirmed, primary-sourced appearance of Grok 4.6 anywhere, after the model had circulated only as an unconfirmed rumor in earlier tracking — it was never found announced on xAI’s own official channels.
Details
- What it’s built for: GitHub describes Grok 4.6 as designed for “agentic coding and complex multi-step workflows.” Internal testing found it performs strongly on terminal-based coding tasks in VS Code and Copilot CLI, particularly on “longer-horizon tasks requiring sustained reasoning and tool use”
- Pricing: Usage-based billing at provider list pricing; see GitHub Copilot’s models-and-pricing documentation for specifics
- Plan availability: Copilot Pro, Pro+, Max, Business, and Enterprise
- Where it works: Visual Studio Code, Visual Studio, Copilot CLI, the GitHub Copilot cloud agent, the GitHub Copilot app, JetBrains IDEs, Xcode, and Eclipse
- Off by default for organizations: Business and Enterprise admins must explicitly enable the Grok 4.6 policy in Copilot settings before members can use it
- Rollout status: Gradual, so it may not appear for every eligible user immediately
What happened next
This GitHub changelog post effectively resolves a loose thread from earlier AI-industry tracking: Grok 4.6 had been referenced as a possible upcoming xAI release but had never turned up on x.ai’s own domain or any other primary xAI source, leaving its status unconfirmed. Its arrival here — as a third-party integration inside GitHub Copilot rather than an xAI announcement — is the first solid, sourceable evidence that the model exists and has shipped, even though xAI itself still hasn’t published anything about it directly.