GitHub Copilot pricing explained

The pair-programmer that moved to usage-based AI Credits (1 credit = $0.01) in June 2026.

The plans

Why the bill surprises people

Picking a frontier model (Claude Opus, GPT-5.5) burns the small $15 Pro credit pool in a few token-heavy agent runs, after which usage continues at full per-token rates against your budget. The launch blog and plans page even disagree on the included-credit amounts, so the numbers are still settling.

How credits map to real token cost

As of June 1 2026, Copilot meters premium-model use as GitHub AI Credits where 1 credit = $0.01, consumed by token usage at each model’s published API rate. Code completions stay unlimited and unbilled; only agent and chat use against premium models draws down credits.

GitHub Copilot runs on Claude Sonnet & Opus, GPT-5. To see what your actual prompts cost at the API level — and which model is cheapest for your workload — paste one into the calculator.

FAQ

Why is my GitHub Copilot bill higher than expected?

Picking a frontier model (Claude Opus, GPT-5.5) burns the small $15 Pro credit pool in a few token-heavy agent runs, after which usage continues at full per-token rates against your budget. The launch blog and plans page even disagree on the included-credit amounts, so the numbers are still settling.

How do credits translate to actual cost?

As of June 1 2026, Copilot meters premium-model use as GitHub AI Credits where 1 credit = $0.01, consumed by token usage at each model’s published API rate. Code completions stay unlimited and unbilled; only agent and chat use against premium models draws down credits.

Which models am I paying for on GitHub Copilot?

GitHub Copilot runs on Claude Sonnet & Opus, GPT-5. You can estimate the raw token cost of any prompt against those models in the tokenmath calculator.

Pricing sourced from GitHub Copilot’s official pricing as of 2026-06-23. Platform pricing changes often — verify before you commit.