Billing is worth studying carefully, because GitHub moved Copilot to a usage-based model on 1 June 2026. This page describes how plans, allowances and billing work today.
Plan names, prices, allowances and billing mechanics on this page are taken from GitHub's own documentation and verified on 26 August 2026. This area moves faster than any other part of the product, so treat the linked source as authoritative if it ever disagrees with this page.
The plans#
| Plan | Price | Notes |
|---|---|---|
| Copilot Free | $0 | 2,000 code completions per month, plus an allowance of AI credits. Auto model selection only |
| Copilot Student | $0 | For verified students. Unlimited code completions, and access to the Copilot cloud agent |
| Copilot Pro | $10 / month | |
| Copilot Pro+ | $39 / month | |
| Copilot Max | $100 / month | The highest individual AI credit allowance |
| Copilot Business | $19 / seat / month | Adds content exclusions and organization-wide policy management |
| Copilot Enterprise | $39 / seat / month |
Seven plans
There are seven Copilot plans. Copilot Max sits above Pro+ at $100 per month with the highest individual AI credit allowance, and Copilot Student is a distinct plan rather than a variant of Free — it includes unlimited code completions and access to the Copilot cloud agent.
Two further details worth knowing:
- Verified teachers and maintainers of popular open source projects get Copilot Pro free. There is no separate “teacher plan” — they receive Pro at no cost.
- Copilot is not available on GitHub Enterprise Server. It is a cloud-only product.
AI Credits, not premium requests#
On 1 June 2026 GitHub moved Copilot to usage-based billing. The unit is the GitHub AI Credit, and 1 AI credit = $0.01 USD.
Models are billed per token
Under AI Credits, model usage is priced per token — input, output and cached tokens each have a rate. There are no request multipliers.
Monthly AI credit allowances#
| Plan | AI credits per month |
|---|---|
| Copilot Free | An allowance of AI credits (GitHub does not publish a figure) |
| Copilot Pro | 1,500 — 1,000 base plus 500 flex |
| Copilot Pro+ | 7,000 — 3,900 base plus 3,100 flex |
| Copilot Max | 20,000 — 10,000 base plus 10,000 flex |
| Copilot Business | 1,900 per user |
| Copilot Enterprise | 3,900 per user |
Code completions and next edit suggestions are not billed in AI credits at all. They remain unlimited on every paid plan. Only model calls through chat, agents and similar features consume credits.
GitHub Docs — Billing for organizations and enterprises
The legacy exception#
Some subscribers are still on the old model
Copilot Pro and Pro+ subscribers who were on an existing annual plan could remain on legacy request-based billing after 1 June 2026. Those accounts still get a monthly premium request quota — Pro 300, Pro+ 1,500 — and still pay $0.04 per additional request with model multipliers applied. New subscriptions and monthly plans use AI Credits.
Business and Enterprise have no per-user premium request allotment under the legacy model.
Models#
The supported model lineup is refreshed regularly. Check the supported models reference for the current list before the exam.
Current supported models
The OpenAI lineup is GPT-5 mini, GPT-5.3-Codex, GPT-5.4, GPT-5.4 mini, GPT-5.4 nano, GPT-5.5, and GPT-5.6 in Luna, Sol and Terra variants. Alongside those sit Anthropic's Claude Haiku, Sonnet and Opus families, Google's Gemini 3.x line, xAI's Grok 4.x, Kimi, Microsoft's MAI-Code models, and Raptor mini.
GPT-5.3-Codex is both the base model and the LTS model, designated on 18 March 2026. Business and Enterprise administrators get a 60-day window before a new base model is enabled automatically.
GitHub Docs — Fallback and LTS models
Budget controls#
Because usage beyond the included allowance is billable, GitHub added spending governance: budgets at user, cost center, organization and enterprise level.
Additional usage is enabled by default
For Business and Enterprise, spending beyond the included allowance is on by default. An administrator who wants a hard cap has to set one explicitly. This is exactly the sort of default-behavior detail an exam likes to test.
What actually matters for the exam#
Billing is not a named domain on GH-300. It surfaces inside Domain 2 — Use GitHub Copilot features, where you are expected to know which features belong to which plan, and to understand subscription management through the REST API.
Learn the feature-to-plan mapping rather than memorizing credit allowances. Which plan gets content exclusions? Which gets organization-wide policy management? Which does not get the cloud agent?
| Feature | Available on |
|---|---|
| Agent mode | All plans, including Free |
| Copilot cloud agent | Student and all paid plans — not Free |
| Copilot Spaces | All plans, including Free |
| MCP | All plans, including Free |
| Audit logs | Pro+, Max, Business, Enterprise |
| Content exclusions | Business and Enterprise only |
| Organization-wide policy management | Business and Enterprise only |
If a billing question feels ambiguous
Prefer the answer grounded in capability over the one grounded in a price or a quota. Prices and allowances change; the tier at which a governance feature appears is far more stable, and is what the objectives actually ask you to know.
Related#
- Domain 2: Use GitHub Copilot features — where plan and policy knowledge is examined
- Domain 6: Configure privacy, content exclusions, and safeguards — content exclusions in depth
- GitHub Docs — Plans for GitHub Copilot — the authoritative source