Is GitHub Copilot still worth it? Yes, Pro at $10 a month is worth it if you code most days and keep running into Free’s limits. If you code now and then, stay on Free: it costs $0, needs no credit card and includes 2,000 completions a month.
Prices are the US-dollar list prices on each vendor’s page, before tax, so what you pay may differ by country. We did not run Copilot for this page. Every fact comes from the vendor pages linked here.
Free vs Pro: what you would actually notice
- Completions. Free stops at 2,000 a month. Pro has no monthly cap, and completions never use credits.
- Chat and agents. Free has “limited chat and agent usage”, with no number published. Pro includes $15 of AI Credits a month: $10 fixed plus $5 of “flex”, which GitHub says may change. You can buy more on Pro at $0.01 per credit. You cannot on Free.
- Code review and cloud tasks. Neither is on Free. Pull request reviews on GitHub, and handing work to Copilot in the cloud, start at Pro.
- Model choice. GitHub’s docs say Free gets automatic model selection only. Pro lets you pick. Opus models start at Pro+, $39 a month.
The simple test: if you reach Free’s limit before the month ends, and it annoys you, pay the $10. If you never reach it, you have no reason to pay yet.
Checked November 4, 2026: what has changed
- April 20, 2026. GitHub paused new Pro, Pro+ and Student sign-ups, tightened usage limits and took Opus models out of Pro. That same month it paused Pro trials. Today the plans page offers sign-up on every individual plan and does not mention a trial.
- June 1, 2026. All plans moved to billing by AI Credits used. A flat fee no longer means flat usage: how many credits a request uses depends on the model and the task. GitHub also added Max at $100 a month.
- Yearly billing. The plans page lists monthly prices only. People already on a yearly plan drop to Free when it ends.
- October 28, 2026. GitHub joined its cloud tools into one experience on paid plans. Free keeps Copilot Chat on github.com and GitHub Mobile. Billing for cloud sandboxes begins in November 2026.
Has something else overtaken it?
For some readers, yes. Two alternatives are worth a look, and neither is a straight swap.
Cursor calls itself a coding agent and comes with its own desktop app. Its pricing page now sells Plus at $8 a month (Grok models only) and Premium at $30 (adds cloud agents and AI code reviews). Third-party models are billed on top, at API list price. If you want an agent-first editor and are happy with Grok models, look there.
Claude Code is included in all paid Claude plans. Claude Pro is $20 a month, or $17 a month billed yearly ($200 up front). If your work is mostly long agent tasks and you want Anthropic’s models, our Claude review covers it.
Copilot’s advantages are reach and price. GitHub lists VS Code, Visual Studio, JetBrains IDEs, Xcode, Neovim, Eclipse and Zed, and its reviews live where many teams already work. At $10 it also costs less than Cursor Premium or Claude Pro.
Worth it if / Skip it if
Worth it if:
- You code most days and reach Free’s limits.
- Your team reviews code on GitHub and you want Copilot’s reviews there.
- You want one assistant across several of the editors above.
Skip it if:
- You code now and then. Free is enough.
- You are a verified student. GitHub has a free Copilot Student plan.
- You want Opus models every day. Pro does not include them.
- You need a predictable bill for heavy agent work. Credits run down faster with bigger models and tasks.
The full breakdown
For every plan, limit and price, read our GitHub Copilot review.