Free
$0/mo
2,000 completions, 50 premium req/mo
GitHub Copilot
From in-editor pair programmer to agentic coding platform.
The most widely deployed AI coding assistant. Inline completions, Chat, agent mode, an autonomous coding agent that opens PRs from issues, and agentic code review — integrated across VS Code, JetBrains, Visual Studio, Neovim, and Xcode. Multi-model: GPT-5.4, Claude Sonnet 4.6 and Opus 4.7, Gemini 2.5 Pro, and more.
+Pros
- Deepest GitHub integration — PRs, issues, Actions, code review, and Copilot all in one platform
- Available across the widest IDE surface: VS Code, JetBrains, Visual Studio, Neovim, Xcode, Azure Data Studio
- Async coding agent handles full issue-to-PR workflows autonomously in GitHub Actions
- Copilot CLI brings agent workflows directly to the terminal with specialized sub-agents
- Enterprise-grade: SOC 2, IP indemnity, SSO, RBAC, private model fine-tuning, admin policy controls
- Multi-model: GPT-5.4, Claude Sonnet, Gemini 2.5 Pro — user-selectable per session
- Free tier available with 2,000 inline suggestions and 50 premium requests per month
−Cons
- IDE extension approach — no standalone AI-native editor, relies on host IDE
- Agent mode less autonomous than Cursor or Claude Code on complex multi-step tasks
- Individual plan sign-ups paused (Apr 2026); moving to usage-based billing from June 2026 — pricing model in transition
- Code sent to Microsoft/GitHub servers — residual privacy concerns for sensitive enterprise codebases
- Experience can feel fragmented across CLI, IDE extensions, Copilot Workspace, and the web interface
Pricing
Pro
$10/mo
Unlimited completions, agent mode
Pro+
$39/mo
Extended models, GitHub Spark
Business
$19/user/mo
Team management, policy controls
Enterprise
$39/user/mo
SSO, audit, IP indemnity, custom models
GitHub Copilot is the most widely used AI coding tool, serving millions of developers across every major IDE. Originally launched as an autocomplete tool in 2021, it has grown into a full agentic platform with chat, agent mode, an autonomous coding agent, agentic code review, and natural-language app building via GitHub Spark.
Inline completions remain context-aware across files, imports, and project conventions. Copilot Chat provides interactive AI assistance in-IDE and on GitHub.com. Agent mode — generally available on both VS Code and JetBrains since March 2026 — lets the agent edit multiple files, run commands, and iterate. The coding agent goes further: assign it a GitHub issue and it works asynchronously in a sandbox until a draft PR is ready. Multi-model access spans GPT-5.4, Claude Sonnet 4.6, Claude Opus 4.7, and Gemini 2.5 Pro.
Key Features
• Inline completions: real-time ghost-text suggestions in every major IDE
• Copilot Chat: conversational assistance in-IDE and on GitHub.com
• Agent mode: multi-step task execution (VS Code + JetBrains)
• Coding agent: autonomous issue-to-PR worker that runs in a sandbox
• Agentic code review: gathers project context, can hand off fixes to the coding agent
• GitHub Spark: natural-language app builder (Pro+ and Enterprise)
• Multi-model: GPT-5.4, Claude Sonnet 4.6 and Opus 4.7, Gemini 2.5 Pro, and more
• Enterprise: SSO, policy controls, audit logs, IP indemnity, custom models
Version History
Copilot CLI reached General Availability (Feb 2026) with terminal-native agent workflows, specialized sub-agents (Explore, Task, Plan, Code-review), parallel execution, and MCP support. GPT-5.4 rolled out across all surfaces (Mar 2026). Copilot SDK launched in public preview for embedding Copilot's agent runtime into custom applications (Apr 2026).
Asynchronous coding agent promoted to General Availability for all paid subscribers. Agent takes GitHub issues, works autonomously in a secure cloud environment (GitHub Actions), and opens pull requests with the completed implementation.
Agent mode launched in VS Code preview: multi-file edits, terminal command execution, self-healing error fixing, and next edit suggestions. Copilot Edits reached GA. Gemini 2.0 Flash and OpenAI o3-mini added as selectable models.
Signature Snippet
Assign a GitHub issue to Copilot: it analyzes the repo, edits files on a branch, runs the test suite in a sandbox, and opens a draft PR with a written summary. Review, comment, and either merge or send it back for another pass.More in this Space
Prime Agent (Prime Intellect)
Prime Agent is an open-source (MIT) self-improving coding and research harness by Prime Intellect, built on a Recursive Language Model that runs inside a persistent IPython REPL and a Continual Harness that refines its own state via /refine. It took the same frontier weights from 30.2% to 95.5% on ARC-AGI-3 RHAE Best@1, powered daemon-backed sessions, and ships no hosted offering.
OpenCodeReview (ocr)
OpenCodeReview (ocr) is Alibaba's open-source AI code-review CLI. It reads git diffs and drives an LLM agent through a hybrid architecture where deterministic pipelines handle file selection, rule matching, and comment positioning while the agent does the reasoning. Built on two years of internal use at Alibaba scale, it ships a multi-language ruleset and consumes roughly 1/9 of the tokens of general-purpose agents.
Delta (by Zed)
Multiplayer environment from Zed for coding with agents. Threads unify the agent conversation, the code edits, and human review in one shareable artifact, replacing the branch/PR loop. Built on DeltaDB, a CRDT-based version control layer that records every edit between commits alongside the prompts and reasoning that produced them.