GitHub Copilot
The widely available AI coding companion built around GitHub.
About
GitHub Copilot is an AI coding assistant that works inside popular editors, in the terminal and on GitHub itself. It suggests code as you type, answers questions about your project in chat, and can carry out multi-file changes in agent mode. Its strongest appeal is breadth: one subscription covers many editors and a deep connection to repositories, pull requests and code review.
Key features
- Inline code suggestions and next-edit suggestions as you type, unlimited on paid plans
- Chat inside supported editors, on GitHub and on mobile, for explanations, refactoring and debugging help
- Agent mode that can plan and apply changes across several files while you review the result
- A command-line assistant for terminal work and a cloud agent that can work from issues and pull requests
- A choice of leading AI models on paid plans, with automatic model selection on the free plan
- Pull-request code review support and tools for generating tests and documentation
- Business and enterprise controls including central management, policies and shared usage pools
Pricing
Pro plan $10/month; Pro+ $39/month; Max $100/month; Business $19 per user/month; Enterprise $39 per user/month
Prices change — check the official pricing page. Chat and agent use draws on included AI credits, while completions do not use credits on paid plans.
Best for
Developers and teams already working on GitHub who want AI help in almost any editor.
Limitations to know
AI code suggestions can be wrong, insecure, or outdated, and they may miss the wider context of a large project. Human review and testing are still required before shipping code.
What users like
- Ease of use and smooth fit into everyday coding workflows, especially in VS Code
- Faster handling of repetitive work such as boilerplate, tests and documentation
- Helpful explanations and suggestions for learning a new codebase or language
What users complain about
- Suggestions that are inaccurate, generic or off target, particularly on complex or unusual code
- Limited understanding of full project context on some tasks
- Confusion and frustration around the newer credit-based billing for chat and agent work
Our editorial score
Our editorial score: 8.5/10 — our own judgement, not an aggregate of user ratings. See the full review for the reasoning.
Read the full GitHub Copilot review · More Coding tools · All AI tools