AI Agent Index
ByHeather MacAvelia·Independently reviewed·Published Apr 23, 2026·Updated Aug 31, 2026
Independently verified against live vendor data on Aug 11, 2026.

Open-source AI coding agent for terminal, IDE, desktop and CI, used by 16M developers monthly. Free and MIT-licensed with BYOK, GitHub Copilot, or ChatGPT Plus/Pro login. SOC 2 Type II.

How we scored it

Autonomy

5/5

Integrations

5/5

Pricing clarity

4/5

Evidence

5/5

Setup

5/5

The facts

OpenCode is the open-source AI coding agent built by Anomaly Innovations, Inc and used by over 16 million developers monthly. With 203k GitHub stars, 900+ contributors, and 14,000+ commits, it is one of the most actively developed AI coding tools available. OpenCode runs in the terminal, as a desktop application (macOS, Windows, Linux, currently in beta), and as an IDE extension. Its defining architecture is model and account flexibility. Developers can connect any of 75+ LLM providers via Models.dev including local models, or use BYOK with API keys from Claude, OpenAI, Gemini, or any OpenAI-compatible endpoint. They can also log in with an existing GitHub Copilot or ChatGPT Plus/Pro account to reuse those subscription quotas without paying for separate API access. The core agent capabilities include LSP-enabled context that automatically loads the right language servers so the LLM understands the codebase's types and imports, plus multi-session parallel execution that runs multiple OpenCode agents simultaneously on the same project. OpenCode ships permissive by default: the Build agent runs file edits and shell commands without per-step approval, while the restricted Plan agent asks before every edit and command, and an --auto flag clears the two remaining guards on external directories and repeated calls. Additional features include share links for session debugging, plan mode for read-only analysis, and undo/redo for iterating on AI-generated edits. OpenCode fully supports the Model Context Protocol as a client for connecting external tools: both local and remote MCP servers are supported, including OAuth authentication for remote servers such as Sentry and Context7. The agent itself is free and MIT licensed, and three optional paid products sit alongside it. OpenCode Go is a subscription at $5 for the first month and $10 per month thereafter, giving access to a curated set of open coding models, with usage capped at $12 per five hours, $30 per week and $60 per month, and one subscriber per workspace. OpenCode Zen is a curated model marketplace billed as a prepaid pay-as-you-go balance starting at $20 plus a $1.23 card processing fee, usable with any agent and cancellable at any time. OpenCode Enterprise adds centralized configuration, SSO and internal AI gateway integration, and is sold per seat on a custom quote rather than a published rate. None of the three is required: the agent runs fully featured on BYOK or a Copilot or ChatGPT login at no cost. Anomaly Innovations, Inc holds a SOC 2 Type 2 report, published through its Comp AI Trust Center at trust.opencode.ai alongside 25 policies and a subprocessor list, with documents available under NDA. It publishes no ISO 27001, HIPAA, FedRAMP or PCI certification, and states no GDPR or CCPA position on any public surface, though a Data Processing Agreement exists in the Trust Center behind an NDA request. OpenCode explicitly does not store code or context data, with all processing happening locally or through direct API calls to the developer's own provider. OpenCode is not the right fit for developers who want a managed subscription with built-in usage quotas and no configuration. Cursor ($16/mo billed annually) and Claude Code ($17/mo billed annually) provide self-contained experiences with billing, support, and model access handled by a single vendor. OpenCode requires the developer to manage their own API keys or subscription accounts, which adds initial setup friction and ongoing billing complexity when usage spans multiple model providers. OpenCode runs inside CI itself: mentioning /opencode or /oc in a GitHub issue or pull request executes the task on your GitHub Actions runner, and GitLab is supported through CI/CD pipelines or GitLab Duo. Teams that want that workflow packaged with seat management, admin controls and vendor support rather than configured per repository should evaluate GitHub Copilot ($10/mo). Current state Q3 2026: OpenCode reached 203k GitHub stars with active daily commits. The desktop app remains in beta on macOS, Windows and Linux. OpenCode Go launched as a low-cost model subscription aimed at international users, and OpenCode Data publishes a public leaderboard of model usage and cost drawn from Go traffic. The platform supports 75+ LLM providers and both local and remote MCP servers with full OAuth support.

Pricing

free

View pricing ↗

Segment

b2b

Setup

easy

Verified

Aug 11, 2026

Transparency

Mostly Public

Contract

Data training

Not Trained

Human in loop

Optional

Capabilities

agentic-codingmulti-file-editingcode-generationterminal-agentopen-sourcebyokgit-native

Pros & Limitations

Editorial assessment

Pros

  • Multi-account architecture lets developers log in with GitHub Copilot or ChatGPT Plus/Pro to reuse existing subscription quotas: materially better unit economics than BYOK-only alternatives requiring separate API keys, and better than single-provider tools for teams already paying for Copilot.
  • Multi-session parallel agents run simultaneously on the same project, with MCP client support for both local and remote servers including OAuth: developers can connect Sentry, GitHub, Context7, and any other MCP-compatible tool without leaving the terminal workflow.
  • Privacy-first architecture with no code or context storage, an MIT license, and a SOC 2 Type 2 report published through the vendor's Trust Center: rare among free developer tools and enough to clear many procurement checks that open-source projects normally fail.

Limitations

  • Permissive by default is a real operational risk: OpenCode ships with most permissions set to allow, so the Build agent edits files and runs shell commands with no approval prompt until an operator writes a permission config, and only repeated identical calls and paths outside the project are gated.
  • Multi-provider flexibility creates setup complexity: developers must manage API keys, billing relationships, and model selection across multiple providers, adding initial friction and ongoing account management compared to single-subscription tools like Cursor or Claude Code.
  • Enterprise terms are unpublished and compliance evidence is gated: per-seat enterprise pricing is quote-only with no public rate, the Trust Center releases its SOC 2 report and DPA only under NDA, and the vendor publishes no ISO 27001, HIPAA or FedRAMP certification.

Technical Details

Deployment
clidesktopide
Model architectureMulti-model (third-party LLMs)
Autonomous ratePermissive by default: OpenCode ships with most permissions set to allow, so the Build agent edits files and runs shell commands without per-step approval. The restricted Plan agent asks for all edits and bash commands
Integrations
GitHub ActionsGitLab CIGitLab DuoVS CodeCursorWindsurfZedJetBrains IDEsNeovimGitHub CopilotChatGPT Plus/ProModels.devSentryContext7Grep by Vercel
Security
SOC 2 Type II

Similar agents

Rating

5.0/ 5

Editorial score

How we score this →

Recognition

Rated 5.0 ★Category Leader · Q3 2026Listed 2026
Embed these on your site →

Industries

B2BEnterpriseSaaSDevToolsOpen Source

Leave a review

Never displayed publicly.

Agent Stacks

See workflow stacks that feature OpenCode.

Compare

Related Content

Is this your tool?

Claim this listing free to get your embeddable badges, a review collection link, and the Verified checkmark.

Claim listing →

Free · Every Two Weeks

AI Agent Price & Rating Tracker

Price changes, new agent launches, acquisitions, and rating updates across the AI agent market. Verified against live vendor data, not vendor marketing.

No spam. Unsubscribe anytime. We never share your email.