opencode

Name: opencode
Rating: 5.0 (39 reviews)

Your terminal's AI agent, with any model you want

5.0•39 reviews•

1K followers

Your terminal's AI agent, with any model you want

5.0•39 reviews•

1K followers

•

•

opencode is an open-source AI coding agent built for the terminal. It features a native TUI, multi-session support, and is model-agnostic, letting you use Claude, OpenAI, Google, or even local models.

Zendesk for StartupsUp to 2 years free for startups + AI Agents & Copilot

Promoted

opencode Launches

opencode zenAI models recommended for developers by the opencode team

Launched on September 29th, 2025

opencodeYour terminal's AI agent, with any model you want

Launched on July 11th, 2025

5.0

Based on 39 reviews

Review opencode?

Reviewers describe opencode as a fast, straightforward terminal coding tool with a clean TUI and unusually flexible model choice, including free and third-party options. Users say it feels responsive, helpful, and not overly intrusive, and one notes editor integration with VS Code, Cursor, and Windsurf. A few drawbacks come up: configuration via jsonc can feel cumbersome, and one reviewer says newer areas like large-codebase navigation, context handling, and agent safety controls still need polish. Founder feedback mostly reinforces its fit in real developer workflows and multi-model setups.

+36

Summarized with AI

Reviews

All Reviews

Most Informative

Athena

•1 review

What's great

Model-agnostic design is a strong differentiator, being able to swap between Claude, OpenAI, Gemini, or local models without lock-in is genuinely useful
Native TUI feels right for a terminal tool; it respects the environment it lives in rather than fighting it
Multi-session support is a practical win for developers juggling multiple contexts
Open-source means the community can audit, extend, and self-host it

What needs improvement

The tool is still relatively new, so context window management and large codebase

navigation likely need more polish

No mention of agentic loop controls (e.g., approval modes, auto-apply guardrails) trust and safety mechanisms matter a lot for an agent that writes code
Terminal-first UX limits accessibility; there's no IDE extension path mentioned, which narrows the audience
"Model-agnostic" is only as strong as the abstraction layer, if switching models silently degrades quality (e.g., tool-use inconsistencies), users may not notice until it's a problem

Ratings

Ease of use

Reliability

Value for money

Customization

Report

203 views5mo ago

•1 review

What's great

cost-effective (1)good context management (1)good ui (1)

It's a good tool for AI coding. I like that it's not too complicated and that it offers free models, while also allowing users to integrate models from other providers.

What needs improvement

configuration complexity (1)

Configuration is a bit cumbersome with jsonc files, but it's not too bad.

vs Alternatives

Codex by OpenAI

Claude Code

The system prompt does not contain any major distortions and is not intended to alter the models' identities or capabilities.

Ratings

Ease of use

Reliability

Value for money

Customization

Report

34 views27d ago

•1 review

I’ve just tried out opencodes zen recently, it’s one of the simplest and fastest setups I’ve come across in a while, for working with LLMs in the terminal. The "TUI" feels super clean and intuitive, no fussing around, just install, configure and start using it. (waiting for its support on the Bun runtime though)

Report

203 views12mo ago

Athena

•1 review

What's great

Model-agnostic design is a strong differentiator, being able to swap between Claude, OpenAI, Gemini, or local models without lock-in is genuinely useful
Native TUI feels right for a terminal tool; it respects the environment it lives in rather than fighting it
Multi-session support is a practical win for developers juggling multiple contexts
Open-source means the community can audit, extend, and self-host it

What needs improvement

The tool is still relatively new, so context window management and large codebase

navigation likely need more polish

No mention of agentic loop controls (e.g., approval modes, auto-apply guardrails) trust and safety mechanisms matter a lot for an agent that writes code
Terminal-first UX limits accessibility; there's no IDE extension path mentioned, which narrows the audience
"Model-agnostic" is only as strong as the abstraction layer, if switching models silently degrades quality (e.g., tool-use inconsistencies), users may not notice until it's a problem

Ratings

Ease of use

Reliability

Value for money

Customization

Report

203 views5mo ago

•1 review

What's great

cost-effective (1)good context management (1)good ui (1)

It's a good tool for AI coding. I like that it's not too complicated and that it offers free models, while also allowing users to integrate models from other providers.

What needs improvement

configuration complexity (1)

Configuration is a bit cumbersome with jsonc files, but it's not too bad.

vs Alternatives

Codex by OpenAI

Claude Code

The system prompt does not contain any major distortions and is not intended to alter the models' identities or capabilities.

Ratings

Ease of use

Reliability

Value for money

Customization

Report

34 views27d ago

•1 review

Report

203 views12mo ago