GitHub Copilot: GPT-6 Sol Now Available

GitHub CopilotView original changelog

GitHub Copilot added GPT-6 Sol, OpenAI's balanced mid-tier GPT-6 model, to its model picker as a general-purpose option for interactive and agentic coding. OpenAI says Sol makes about half as many mistakes as its GPT-5.6 predecessor while costing roughly half as much per token, thanks to caching and inference improvements rather than a smaller model. It's available to Copilot Pro+, Max, Business, and Enterprise subscribers under usage-based billing.

Featured Video

A video we selected to help illustrate this changelog

Key Takeaways

  • A YouTube review titled "GPT-6 Sol Is Here (50% Cheaper!)" highlights the model's headline change: roughly half the per-token price of its GPT-5.6 Sol predecessor.
  • The video walks through Sol's positioning as the balanced, all-round coding model in the GPT-6 lineup, sitting between the flagship Astra and the budget Luna.
  • OpenAI says Sol makes about half as many mistakes as GPT-5.6 Sol, approaching GPT-6 Astra-level reliability at a fraction of the cost.
  • The price cut comes from caching and inference improvements, not a smaller or less capable model, according to the video and OpenAI's own release notes.
  • GPT-6 Sol is priced at $2 input / $10 output per million tokens, and is available to Copilot Pro+, Max, Business, and Enterprise subscribers.
  • TechCrunch and VentureBeat both frame the Sol and Luna launch as part of OpenAI's push to undercut rivals on price across its smaller, faster models.

GPT-6 Sol Joins GitHub Copilot's Model LineupGitHub Copilot added GPT-6 Sol, OpenAI's balanced mid-tier model from the GPT-6 family, to its model picker alongside the previously released GPT-6 Astra. GitHub describes Sol as a strong all-round choice for development tasks that benefit from careful, multistep validation, positioning it as the everyday option for interactive and agentic coding rather than the flagship or the budget tier.

Fewer Mistakes at Half the PriceOpenAI's own announcement, corroborated by TechCrunch and VentureBeat coverage, says GPT-6 Sol makes about half as many mistakes as its GPT-5.6 Sol predecessor while approaching GPT-6 Astra-level reliability, all at roughly half of GPT-5.6 Sol's per-token price. OpenAI attributes the price cut to improvements in caching and inference efficiency rather than a smaller or weaker model.

AvailabilityGPT-6 Sol is available to Copilot Pro+, Max, Business, and Enterprise subscribers under usage-based billing, selectable across Visual Studio Code, Visual Studio, Copilot CLI, the GitHub Copilot cloud agent, the GitHub Copilot app, github.com, GitHub Mobile, JetBrains IDEs, Xcode, and Eclipse. Rollout is gradual, and Enterprise and Business administrators can control access through Copilot's model policy, where new models are enabled by default unless explicitly turned off.