OpenCode
FreeThe open source AI coding agent
OpenCode is an open-source AI coding agent that helps developers write, understand, and modify code through a terminal interface, desktop app, or IDE extension. It connects to more than 75 LLM providers through Models.dev and supports local models, so users can bring their own API keys or run code entirely on their own machines. The project is released under the MIT license and is built around a privacy-first architecture that does not store user code or context.
What is it
OpenCode is an open-source agent that assists with coding tasks across the terminal, desktop, and IDE extension interfaces.
What it can do
It reads codebases, answers questions about code, plans changes in a read-only mode, edits files, runs shell commands, supports multiple simultaneous sessions, and connects to a wide range of LLM providers or local models.
Who is it for
Software developers, engineering teams, and privacy-conscious users who want a transparent, provider-agnostic coding assistant that runs locally.
Key Features
Multi-Provider Model Support — Connect to 75+ providers or local models
OpenCode integrates with providers such as Claude, GPT, and Gemini through Models.dev, and supports local models via Ollama and other OpenAI-compatible servers. Users can choose models based on capability, cost, or privacy requirements without being locked to a single vendor.
Build and Plan Agents — Switch between full-access and read-only modes
The default Build agent can edit files and run commands, while the Plan agent analyzes codebases and proposes changes without modifying files. A General subagent is also available for complex multi-step searches.
LSP Integration — Automatic language server loading for the selected model
OpenCode loads the appropriate Language Server Protocol servers so the agent receives compiler and linter feedback after edits, helping it catch type errors and other issues in real time.
Multi-Session Support — Run parallel agents on the same project
Users can start multiple independent sessions against one codebase, each with its own conversation history and focus area.
Use Cases
Explore unfamiliar codebases
Developers switch to Plan mode and ask OpenCode to explain architecture, dependencies, or authentication flows before making changes.
Add features across multiple files
Engineers describe a feature in natural language and let the Build agent plan, edit, test, and iterate until the implementation passes local checks.
Debug and fix production issues
Users point OpenCode at failing code, ask it to identify the root cause, and review the suggested edits and test runs before applying them.
Refactor legacy code
Teams ask OpenCode to restructure services, replace conditional chains with patterns, or modernize syntax while preserving existing behavior.
Pricing plans
Free to use
This tool is listed as free. There is no paid pricing page to show here—visit the official site for any usage limits, quotas, or terms of service.
Frequently Asked Questions
Discussion
No comments yet. Be the first to start the thread.