| | |

GPT-6.1 Terra in GitHub Copilot: Features, Plans and Cost

Developer typing at computer monitors illustrating GPT-6.1 Terra coding workflows.

GPT-6.1 Terra is now available in GitHub Copilot. GitHub announced general availability on October 26, 2026, describing OpenAI’s new model as a fast, efficient option for everyday agentic coding and terminal tasks. Separately, OpenAI says Terra is built for visual reasoning and precise computer use. For developers, the useful question is not simply whether a model is new: it is where you can select it, what it costs and which tasks suit it.

Last checked 3 November 2026 | BigExclusives Editorial | Based on official GitHub and OpenAI product announcements. No change to Copilot access or pricing has been announced since October 26, 2026.

What is GPT-6.1 Terra?

GPT-6.1 Terra is a model in OpenAI’s GPT-6.1 lineup, designed to balance capability, speed and cost. GitHub positions its Copilot integration around focused agentic coding and terminal workflows. OpenAI’s broader announcement highlights the model’s ability to interpret visual information and perform computer-based tasks, while producing clear written output.

That does not mean it is the best model for every software problem. OpenAI’s published guidance continues to distinguish lower-cost, focused work from particularly difficult reasoning and engineering tasks. The right selection depends on your task, accuracy needs and budget.

Which GitHub Copilot plans support GPT-6.1 Terra?

GitHub’s October 26 changelog lists access for the following Copilot plans:

  • Copilot Pro
  • Copilot Pro+
  • Copilot Max
  • Copilot Business
  • Copilot Enterprise

GitHub says the rollout is gradual. An eligible subscriber may not see Terra immediately. Business and Enterprise administrators can also control model access through organizational policies.

Where can developers select the new model?

GitHub lists support through the model picker in Visual Studio Code, Visual Studio, GitHub Copilot CLI, GitHub’s cloud coding agent, the Copilot app, github.com, GitHub Mobile, JetBrains IDEs, Xcode and Eclipse. The exact menu and policies may differ by client and account.

Try this simple workflow:

  1. Update or open your supported Copilot client and sign in with an eligible subscription.
  2. Open the available model picker and look for GPT-6.1 Terra.
  3. If it does not appear, check your organization model policy and GitHub’s rollout notices rather than assuming a broken installation.
  4. Start with a contained coding task such as writing tests, explaining a function, improving error messages or changing one component.
  5. Review the proposed changes and run the real test suite before merging or deploying anything.

How much does GPT-6.1 Terra cost?

GitHub says the model is billed at provider list pricing under its usage-based billing approach. Your actual Copilot charges and included usage depend on GitHub’s current pricing and subscription rules; being listed in a plan does not necessarily mean unlimited free requests.

For developers accessing the model directly through OpenAI’s API, the official OpenAI announcement lists standard prices of $1 per million input tokens, $0.05 per million cached input tokens, and $5 per million output tokens. These are public API token prices, not a substitute for checking GitHub Copilot’s own billing.

A basic cost check should consider prompt size, output length, tool calls and the number of revisions. For high-volume tasks, inexpensive individual requests can still add up, so compare verified cost per completed task rather than token price alone.

Update — October 31, 2026: five days after launch, the position is unchanged — and the wider rollout picture is now clearer. Terra shipped on October 26, 2026 across the OpenAI API (as gpt-6.1-terra), ChatGPT Work and Codex for paid plans, GitHub Copilot, Amazon Bedrock, and Microsoft’s Azure Foundry model offering, so developers can adopt it without changing platforms. OpenAI’s pricing is $1 per million input tokens and $5 per million output tokens — which OpenAI presents as 50% below both GPT-5.6 Terra and GPT-6.1 Sol — and the model is positioned as the mid-tier option between the cheaper Luna models and the more capable Sol and Astra models. GitHub’s rollout in Copilot remains gradual, so an eligible subscriber who still cannot see Terra in the model picker should check organisation model policy and GitHub’s rollout notices rather than assume a fault. One date to note: GitHub has scheduled GPT-5.6 Terra — the model GPT-6.1 Terra replaces — for retirement on November 16, 2026, so anyone still on the older Terra model should trial the new one before that date. [https://explainx.ai/blog/gpt-6-1-terra-launch-1-5-dollar-pricing-computer-use-benchmarks-2026] [https://metarix.io/gpt-6-1-terra-pricing] [https://pondero.ai/coding/guides/github-copilot-model-deprecation/]

What is the difference between Terra and a more powerful model?

GitHub emphasizes focused tasks, terminal operations and efficiency for Terra. More demanding work—for example, redesigning an architecture, auditing complex authorization flows or coordinating large multi-file refactors—may call for a stronger reasoning model, stricter verification or both. Vendor benchmark statements are useful background but are not a replacement for testing on your own project.

For a longer engineering workflow with security and deployment checks, see our AI coding workflow guide and the no-code builder vs AI coding comparison.

Does GPT-6.1 Terra also work in ChatGPT?

OpenAI’s October 2026 documentation describes Terra in ChatGPT Work and Codex, subject to availability by account, plan, region and workspace controls. Do not assume the same model is present in an ordinary Chat conversation simply because it is available in Work or a coding tool.

See the ChatGPT Learn release notes for current product placement and usage guidance.

A responsible five-minute test before using AI-generated code

  • Give the model a narrowly scoped change and an acceptance test.
  • Inspect the diff; ensure it did not change unrelated files.
  • Run unit tests, linting and a production build.
  • Review dependencies, secrets and user-input handling.
  • Compare completion time, developer corrections and actual cost against your previous workflow.

Speed is valuable only when the resulting application still works safely. For deployment checks, our website backup guide covers preparation for reversible changes.

Frequently asked questions

Is GPT-6.1 Terra free in GitHub Copilot?

GitHub lists supported subscription tiers and usage-based billing. Do not assume it is included without limits; check your plan’s current model and billing page.

Why can’t I see Terra in Copilot?

The rollout is gradual. Client compatibility, subscription eligibility and administrator policy can all affect availability.

Is GPT-6.1 Terra an API model?

Yes. OpenAI documents the API model identifier as gpt-6.1-terra.

Should I use Terra for a production security audit?

You may use AI to assist an audit, but qualified review, real tests and independent verification remain necessary. No model should be treated as an automatic production approval.

Official sources

Editorial transparency: Features, availability and prices are from published provider materials, not an independently conducted benchmark. Confirm today’s terms in the provider interface before subscribing.

Similar Posts