GitHub Copilot

GitHub Copilot started as an in-editor autocomplete tool, but the current product is a full multi-model coding platform. It lives inside your editor, in the terminal, in a desktop app, and on GitHub itself, handling code completion, chat, an agent mode that edits across files, cloud agents that work on tasks asynchronously, and automated code review.

GitHub has repositioned Copilot around a new desktop app pitched as taking issues all the way to merge, and billing now runs on consumption-based GitHub AI Credits rather than flat request units. Paid plans include both a base monthly credit allowance and an optional flex allowance, and higher tiers can now delegate work to third-party coding agents such as Claude by Anthropic and OpenAI Codex, currently in preview.

The core capabilities are agent mode, cloud agents, chat, code review, a CLI, and the ability to pick from models across multiple vendors including Anthropic, OpenAI, Google, xAI, Microsoft and Moonshot. Inline code completions and Next Edit suggestions stay included on every plan and do not draw down your AI Credits allowance.

On pricing, individuals can choose Free at $0 per month (about 2,000 inline suggestions, limited chat and agent usage, no code review or cloud agents), Pro at $10 per month in base credits plus $5 per month in flex credits for $15 per month total (1,500 AI Credits, unlimited inline suggestions, code review and cloud agents), Pro+ at $39 per month base plus $31 per month flex for $70 per month total (3,900 plus 3,100 credits, adds delegation to third-party coding agents), or Max at $100 per month base plus $100 per month flex for $200 per month total (10,000 plus 10,000 credits, priority access to premium models). A separate Copilot Student plan gives verified students their own AI Credits allowance, and verified teachers and maintainers of popular open source projects can get Pro for free. Organizations pick Business at $19 per user per month (1,900 AI Credits per user) or Enterprise at $39 per user per month (3,900 AI Credits per user, requires GitHub Enterprise Cloud).

The main limitation is cost predictability: usage-based AI Credits billing, split into base and flex allotments, means heavy agent, chat or premium-model use can push real monthly cost well past the advertised base price. Admin budgets and spending caps help organizations control this, but teams still need to monitor consumption closely.

Copilot is a strong fit if you already work in GitHub and VS Code, want an agentic assistant rather than plain autocomplete, and value being able to choose from models across different vendors. The free tier is a low-risk way to try it, while Pro+ and Max suit heavier users who want premium models and third-party agent delegation.