OpenCode

Open-source, model-agnostic terminal coding agent — bring your own API keys.

Open Source Coding & Development
Visit Website ↗ by Anomaly

What is OpenCode?

OpenCode is an open-source (MIT) terminal-first AI coding agent from the team behind SST (now Anomaly), and one of the most popular open-source coding agents by GitHub stars. It runs as a rich terminal UI on your machine and connects to 75+ model providers — Anthropic, OpenAI, Google, Bedrock, Groq, OpenRouter, and local models via Ollama — so you bring your own keys and pay only for tokens. LSP integration feeds real compiler diagnostics back to the model, sessions support undo/redo via git snapshots, and there is a plan mode, multi-session parallel execution, background subagents, MCP support, and optional desktop and IDE-extension surfaces.

Key features

  • Terminal TUI with multi-session parallel execution
  • 75+ model providers via bring-your-own keys
  • LSP integration with compiler diagnostics in context
  • Plan mode, git-based undo/redo
  • MCP client for local and remote servers
  • Desktop app and IDE extensions
  • Background subagents and share links
  • Air-gapped deployment with local models

Watch: OpenCode tutorial

OpenCode Full Tutorial: Free Models, Skills & MCPs OpenCode Full Tutorial: Free Models, Skills & MCPs — Eric Tech

Video by Eric Tech on YouTube — plays here without leaving this page.

Pros

  • Fully open source (MIT) and model-agnostic
  • Cheap at scale: no subscription markup on tokens
  • Runs fully offline with local models

Cons

  • Configuration overhead to get the most out of it
  • You manage your own API keys and costs
  • Very fast release cadence can introduce rough edges

OpenCode at a glance

PricingOpen Source
CategoryCoding & Development
Platformswindows, macos, linux
CompanyAnomaly
Official websiteopencode.ai
coding-agentcliopen-sourcebyokmulti-modelmcp

How to choose the right Coding & Development tool

The right coding tool depends on where you write code and how much of your codebase it can understand. IDE integration is the first filter — a tool that lives in your editor, sees your project files and respects your stack will help far more than a general chatbot you paste snippets into. Look for accurate completions, clear explanations of suggested changes, and strong refactoring and debugging support for your main languages. Privacy deserves real attention: check whether your code is used for training, and whether a self-hosted or enterprise option exists for sensitive projects. Price against the hours it actually saves — most developers feel the difference within the first week of real use.

Compare all Coding & Development tools →

OpenCode FAQs

What is OpenCode?

OpenCode is an open-source (MIT) terminal-first AI coding agent from the team behind SST (now Anomaly), and one of the most popular open-source coding agents by GitHub stars. It runs as a rich terminal UI on your machine and connects to 75+ model providers — Anthropic, OpenAI, Google, Bedrock, Groq, OpenRouter, and local models via Ollama — so you bring your own keys and pay only for tokens. LSP integration feeds real compiler diagnostics back to the model, sessions support undo/redo via git snapshots, and there is a plan mode, multi-session parallel execution, background subagents, MCP support, and optional desktop and IDE-extension surfaces.

What are the main features of OpenCode?

OpenCode offers Terminal TUI with multi-session parallel execution, 75+ model providers via bring-your-own keys, LSP integration with compiler diagnostics in context, Plan mode, git-based undo/redo, MCP client for local and remote servers, Desktop app and IDE extensions, Background subagents and share links and Air-gapped deployment with local models.

How much does OpenCode cost?

OpenCode is open source, so the software itself is free to download, use and self-host. Your only costs are the infrastructure you run it on and any paid services you connect to it.

What are the pros and cons of OpenCode?

Strengths of OpenCode include Fully open source (MIT) and model-agnostic; Cheap at scale: no subscription markup on tokens; Runs fully offline with local models. On the downside, Configuration overhead to get the most out of it; You manage your own API keys and costs; Very fast release cadence can introduce rough edges.