OpenCode
Open-source AI coding agent that runs as a terminal UI, desktop app or IDE extension. Works with 75+ model providers through Models.dev (including local models) and can sign in with your ChatGPT Plus/Pro or GitHub Copilot account.
Open-source coding agent · Pricing: Free and open source (MIT), with free models included; you pay any other provider you connect
Pros
- Model freedom: 75+ providers, local models, or your ChatGPT Plus/Pro or Copilot login
- /goal (v2.2.1, Oct 29, 2026) keeps a session working until a reviewer step confirms the objective is met
- Built-in Claude Agent provider (v2.2.3) lets you use your Claude Code login inside OpenCode
- MIT licensed, with parallel multi-session agents and plugins
Cons
- Very fast release cadence with frequent breaking changes for plugin and SDK authors
- Scheduled prompts, image generation and artifacts are still opt-in experimental features
- Goal reviews run extra model calls, so they use more of your model budget
Recent changes
- 2026-10-29 OpenCode 2.2.1 adds /goal: set an objective and the session keeps working until a reviewer step confirms it is met, with pause, resume and clear controls, plus opt-in scheduled prompts via /schedule. v2.2.3 (Oct 30) adds a built-in Claude Agent provider that uses your Claude Code login. Source
Ranked in
- Best AI coding tools for an existing codebase (#8)
- Best terminal (CLI) coding agents (#3)
- Best open-source AI coding agents (#1)
Compare
Alternatives to OpenCode
- Claude Code (Terminal coding agent)
- Cursor (AI code editor)
- GitHub Copilot (AI coding assistant)
- OpenAI Codex (Coding agent)
- Devin Desktop (formerly Windsurf) (AI code editor)
- Kiro (Spec-driven coding agent)
FAQ
What is OpenCode?
OpenCode is a open-source coding agent. Open-source AI coding agent that runs as a terminal UI, desktop app or IDE extension. Works with 75+ model providers through Models.dev (including local models) and can sign in with your ChatGPT Plus/Pro or GitHub Copilot account.
How much does OpenCode cost?
Free and open source (MIT), with free models included; you pay any other provider you connect. BuildJury only states prices confirmed on the official pricing page; check https://opencode.ai for current plans.
What is OpenCode best for?
BuildJury ranks OpenCode for: Existing codebase (#8), Terminal coding agents (#3), Open-source coding agents (#1).
What are the downsides of OpenCode?
Very fast release cadence with frequent breaking changes for plugin and SDK authors. Scheduled prompts, image generation and artifacts are still opt-in experimental features. Goal reviews run extra model calls, so they use more of your model budget.