GitHub Copilot

GitHub Copilot is Microsoft-owned GitHub's AI coding assistant, spanning inline completion, chat, autonomous coding agents that open pull requests, AI code review and a command-line agent. It is aimed at engineering organisations that already live inside GitHub and want AI assistance governed by the same policies, audit logs, compliance scope and IP indemnity as their source code.

GitHub Copilot began in June 2021 as an inline code-completion model and has since become an agent platform layered onto the GitHub workflow itself. It now spans four surfaces: inline completions and Copilot Chat inside Visual Studio Code, Visual Studio, JetBrains IDEs, Neovim, Vim, Eclipse, Azure Data Studio and GitHub Mobile; a coding agent that takes an issue, works in the background and returns a pull request; AI code review that can be enabled across all pull requests in an organisation, including from unlicensed users; and Copilot CLI, which in 2026 gained custom agents defined in plain Markdown files so a team can encode its stack conventions, tooling and standards as reusable workflows, with specialised built-in agents such as Explore for codebase analysis and Task for running tests and builds in parallel. GitHub's Agent HQ initiative extends the same paid subscription to third-party agents from Anthropic, OpenAI, Google, Cognition and xAI, surfaced through a mission control view for assigning, steering and tracking parallel agent runs across GitHub, VS Code, mobile and the CLI. Copilot Spaces let teams pin repositories and documentation as durable project context. Model choice is explicit rather than fixed, with a catalogue spanning Claude, GPT and Gemini variants and the most expensive models gated to higher tiers. Billing changed materially on 1 June 2026, moving from Premium Request Units to token-metered GitHub AI Credits pooled across an enterprise; code completions and next-edit suggestions stay unlimited on every paid plan, while agent and premium-model usage draws down the pool and overage bills at $0.01 per credit. GitHub describes millions of individual users and tens of thousands of business customers, which makes Copilot the default incumbent every rival is benchmarked against.

Category

Developer Tools

Pricing

Subscription, Usage-based, Freemium

Target Market

CTOs, CIOs, VPs of Engineering, Enterprise Developers, Platform Engineering Teams

Deployment

Cloud-only

Founded

2008

Headquarters

San Francisco, United States

Team Size

500+

Customers

GitHub states millions of individual users and tens of thousands of business customers

• ✓
  Inline completions and next-edit suggestions

  Unlimited on every paid plan and explicitly excluded from AI-credit metering, so the everyday autocomplete path carries no variable cost.

• ✓
  Copilot coding agent

  Takes an assigned issue, works asynchronously in the background and returns a pull request for human review rather than editing live.

• ✓
  AI code review on pull requests

  Can be enabled organisation-wide across all pull requests, including those from unlicensed contributors, with cost drawn from pooled AI credits.

• ✓
  Agent HQ and mission control

  One command centre to assign, steer and track multiple agents — including third-party agents from Anthropic, OpenAI, Google and xAI — across GitHub, VS Code, mobile and CLI.

• ✓
  Copilot CLI with custom agents

  Markdown-defined custom agents encode a team's stack context, available tools and output standards into reusable terminal workflows that can run in parallel.

• ✓
  Copilot Spaces

  Pins repositories and documentation into a shared, reusable context bundle so Copilot answers as a project expert instead of guessing.

• ✓
  Enterprise governance controls

  Content exclusions, configurable EU data residency, org-wide policy controls, audit logs and IP indemnity, all administered from GitHub Enterprise Cloud.

• •
  Backlog burn-down with background agents

  Assign well-specified issues to the coding agent, which opens draft pull requests for engineers to review instead of writing the first pass by hand.

• •
  Pull-request review at scale

  Turn on AI review across every pull request so the first pass catches routine defects before a human reviewer spends time on the change.

• •
  Boilerplate and scaffolding

  Generate project scaffolding, test cases and repetitive logic inline, which reviewers on Capterra and Product Hunt cite as the most reliable payoff.

• •
  Onboarding to an unfamiliar codebase

  Use Copilot Spaces and Chat to answer questions grounded in the team's own repositories and documentation rather than generic public knowledge.

• •
  Terminal-side automation with custom agents

  Encode stack conventions, test commands and build steps into Markdown-defined CLI agents that any engineer on the team can invoke consistently.

List pricing is fully published. Since 1 June 2026 every paid seat carries a monthly pool of GitHub AI Credits rather than a flat allowance of premium requests: Business is $19/user/month with 1,900 credits, Enterprise $39/user/month with 3,900. Credits pool across the enterprise and overage bills at $0.01 per credit, so agent-heavy teams should model variable spend. Code completions and next-edit suggestions are explicitly not metered and stay unlimited on paid plans, which keeps the everyday autocomplete case at fixed cost. Enterprise is only available on GitHub Enterprise Cloud, and EU data residency plus IP indemnity are the two things that most often force the upgrade from Business.

This page was written from 5 sources, 3 on domains other than github.com.

1. 1.github.com — copilotvendor
2. 2.github.com — plansvendor
3. 3.docs.github.com — organizations and enterprises
4. 4.capterra.com — reviews
5. 5.producthunt.com — reviews