PRICING

GitHub Copilot pricing: what the tiers gate, not just what they cost

checked 2026-11-02
The short answer

Budget by seat, not by token. Copilot's tiers are gated by seats and by admin features rather than by model usage, which is why it is the easiest tool here to put in a spreadsheet. Verify the current list price on the vendor page before you commit — it is the field we could not confirm.

Two axes: individual and organisation

Individual plans sell model access and usage allowances to one person. Organisation plans sell seats and, more importantly, controls: policy management, audit logs, content exclusion and the ability to block suggestions matching public code.

The feature that usually forces the upgrade is not usage — it is audit. Security teams ask who can see what, and that answer only exists above the business tier.

Where the bill actually moves

Three levers: seat count, tier, and whether you buy premium model access. Unlike metered agents, a heavy user does not generate a surprise invoice; they hit a usage limit, which is a very different failure mode.

That predictability is the product. Teams that complain Copilot is expensive are usually comparing it to a metered tool at light usage, which is the one comparison where metering wins.

What to check before renewing

Confirm the current list price and what the free tier includes this quarter — this is the fastest-changing field in the category.

Check whether the features you actually use (agent mode, code review, custom instructions) sit on the individual or the business tier. Buying seats for controls you do not use is the common overpay.

PlanPriceNotes
Free$0Limited quota

FAQ

Why is there no number on this page?

Because we could not verify GitHub's current list prices against a readable pricing page at publish time, and a stale price is worse than an explicit gap. The structure above is stable; the number is not.

Is Copilot cheaper than Cursor?

At light usage, almost always yes — per seat versus subscription plus metered model usage. At heavy agent usage the comparison inverts, because Cursor's usage sits inside the tier while Copilot's agent features are bounded by plan.

Last checked 2026-11-02. Prices and feature gates change; each tool page links to its vendor pricing page so you can confirm before committing.