Not for Teams that want a single vendor's fully-managed, zero-setup hosted service with a traditional support contract, or anyone who needs to run OpenCode's local server on a shared or untrusted network without extra hardening given its disclosed 2026 security history.

Changed since our last check

Runs autonomously: now Yes. It was Partly when we checked on 19 October 2026.
How it meters: now Yes. It was Partly when we checked on 19 October 2026.

The facts, with their sources

What it does

Writes code: YesAuthors new code from a prompt across the terminal, desktop and IDE surfaces; the docs walk through describing a feature and letting it build it.OpenCode is an open source agent that helps you write code in your terminal, IDE, or desktop.opencode.ai
Debugs and fixes: YesCan investigate a reported failure and land the fix: in GitHub it triages an issue, then works in a branch and opens a pull request.Fix and implement: Ask OpenCode to fix an issue or implement a feature. And it will work in a new branch and submits a PR with all the changes.opencode.ai/docs/github
Reviews code: YesReviews diffs others wrote: a documented workflow runs it on pull_request events, and PR review comments hand it the file path, line numbers and diff context.pull_request | PR opened or updated | Automatically trigger OpenCode when PRs are opened, synchronized, or reopened. Useful for automated reviews.opencode.ai/docs/github
Whole-codebase work: PartlyBuilt for multi-step work across a repository through subagents and LSP loading, but the docs describe no repository-wide index or migration guarantee, so scale rests on agent orchestration.A general-purpose agent for researching complex questions and executing multi-step tasks. Has full tool access, so it can make file changes when…opencode.ai/docs/agents
Ships it: PartlyStops at a branch and a pull request in the documented GitHub flow; deployment is only possible because shell commands are permitted, not as a described feature.OpenCode reads context and can create branches, open PRs, or reply.opencode.ai/docs/github
Done-for-you service: NoYou run the software yourself: it installs as a CLI, desktop app or IDE extension and the vendor sells model access, not delivery work on your behalf.OpenCode is an open source AI coding agent. It's available as a terminal-based interface, desktop app, or web app.opencode.ai/v2/docs

How much it does unattended

Runs autonomously: YesAn auto-approve mode clears every permission prompt that is not explicitly denied, and non-interactive runs plus scheduled GitHub workflows execute with no one watching.Was Partly before our last check.Start OpenCode with --auto to automatically approve permission requests that are not explicitly denied.opencode.ai/docs/permissions
Multi-agent: YesPrimary agents invoke subagents that open child sessions, and several agents can be run in parallel on one project from separate sessions.Subagents are specialized assistants that primary agents can invoke for specific tasks. You can also manually invoke them by @ mentioning them in…opencode.ai/docs/agents
Agent permissions: YesEach tool resolves to allow, ask or deny, with glob rules per shell command or file path and per-agent overrides; defaults are permissive apart from external directories.Each permission rule resolves to one of: "allow" — run without approval; "ask" — prompt for approval; "deny" — block the actionopencode.ai/docs/permissions

Which models it runs on

Claude: YesClaude models are available both through an Anthropic API key and through the vendor's own gateway, and a Claude subscription can be connected directly.Yes, OpenCode supports subscription plans from all major providers. You can use your Claude Pro/Max, ChatGPT Plus/Pro, or GitHub Copilot…opencode.ai
GPT: YesGPT models are listed on the vendor's own gateway with per-token prices, and a ChatGPT Plus or Pro login can be used instead of an API key.ChatGPT Plus/Pro Log in with OpenAI to use your ChatGPT Plus or Pro accountopencode.ai
Open models: YesOpen-weight families such as Qwen, DeepSeek, GLM and Kimi are offered on the gateway, and locally served models can be added as a custom provider.OpenCode uses the AI SDK and Models.dev to support 75+ LLM providers and it supports running local models.opencode.ai/docs/providers
Model choice: YesYou pick the model per session or per agent in provider/model form; the vendor's own gateway is explicitly optional rather than a default lock-in.Free models included or connect any model from any provider, including Claude, GPT, Gemini and more.opencode.ai

Where you use it

In your editor: YesAn extension covers VS Code and its forks with keyboard launch, selection sharing and file-reference shortcuts; it installs on first terminal run.OpenCode integrates with VS Code, Cursor, or any IDE that supports a terminal. Just run opencode in the terminal to get started.opencode.ai/docs/ide
On the command line: YesThe terminal is the primary surface: a TUI by default, plus a non-interactive run command for scripting and automation.Run opencode in non-interactive mode by passing a prompt directly.opencode.ai/docs/cli
In your pipeline: YesA GitHub Action runs it inside your own Actions runner, triggered by comments, PR events, a cron schedule or a manual dispatch.OpenCode integrates with your GitHub workflow. Mention /opencode or /oc in your comment, and OpenCode will execute tasks within your GitHub Actions…opencode.ai/docs/github
In a browser: YesA browser interface is served by the local binary, bound to localhost by default and optionally exposed on the network with basic auth.OpenCode can run as a web application in your browser, providing the same powerful AI coding experience without needing a terminal.opencode.ai/docs/web

Whose machine it runs on

Self-hosted: YesThe agent runs on your own machine or server with no vendor backend in the path, and enterprises can point it at an internal AI gateway only.OpenCode does not store your code or context data. All processing happens locally or through direct API calls to your AI provider.opencode.ai/docs/enterprise
Open source: YesSource is public under the MIT licence and the project takes outside issues and pull requests, so the agent itself can be forked and modified.Yes, OpenCode is fully open source. The source code is public on GitHub under the MIT License, meaning anyone can use, modify, or contribute to its…opencode.ai

What it costs to run

How it meters: YesThree meters coexist: the agent is free, the gateway bills per million tokens, a $10 or $40 monthly plan covers usage limits, and enterprise is per seat.Was Partly before our last check.Console uses pay-as-you-go pricing. Below are the prices per 1M tokens.opencode.ai/v2/docs/console/models
Free tier: YesNo call and no account needed to start: the agent is free and ships with free models, with cost arriving only when you connect a paid provider.OpenCode is 100% free to use. It also comes with a set of free models. There might be additional costs if you connect any other provider.opencode.ai
API access: YesA headless server publishes an OpenAPI 3.1 spec used to generate the SDK, covering sessions, messages, files and agents, with optional basic auth.The opencode serve command runs a headless HTTP server that exposes an OpenAPI endpoint that an opencode client can use.opencode.ai/docs/server
MCP server: PartlyMCP runs one way here: OpenCode consumes local and remote MCP servers, and no MCP server of its own is documented, though its HTTP API lets other software drive it.You can add external tools to OpenCode using the Model Context Protocol, or MCP. OpenCode supports both local and remote servers.opencode.ai/docs/mcp-servers
Bring your own key: YesYour own keys are the default path: credentials are stored locally per provider, base URLs are overridable, and enterprises can force an internal gateway.With OpenCode you can use any LLM provider by configuring their API keys.opencode.ai/docs

Buying it for a team

A company can buy it: YesOrganisations buy through a Console workspace with roles, invitations and shared credits, or through a quoted per-seat enterprise agreement.A workspace holds everything in the Console: members, credits, budgets, providers, policies, and service accounts. Usage and billing are tracked per…opencode.ai/v2/docs/console/workspaces
Seat model: PartlyThe gateway charges no seat fee at any team size, so seats only appear in the unpriced enterprise agreement, which is quoted rather than published.We use a per-seat model for OpenCode Enterprise. If you have your own LLM gateway, we do not charge for tokens used.opencode.ai/docs/enterprise
Pooled budget: YesOne prepaid balance covers the whole workspace, with an optional per-member limit layered on top; a request must fit both ceilings.Console is free for any team size, with no seat fees. You pay only for usage, from prepaid credits that cover the whole workspace.opencode.ai/v2/docs/console/billing
Admin controls: YesAdmins constrain spend and model choice in the workspace, and an enterprise central config can restrict the agent to one approved gateway.You can also disable all other AI providers, ensuring all requests go through your organization's approved infrastructure.opencode.ai/docs/enterprise
Audit log: PartlyUsage and request logs per member are recorded and exportable as CSV, but no log of administrative actions such as role or budget changes is documented.Click a member to see their monthly budget, models and clients they use, usage charts, and request logs.opencode.ai/v2/docs/console/members
Single sign-on: YesBoth protocols are supported with named providers, one connection per workspace, and it is included rather than gated behind an enterprise tier.Single sign-on (SSO) sends workspace members through your company's identity provider when they sign in. Owners and admins set it up in Settings ›…opencode.ai/v2/docs/console/sso

What happens to your code

Opt out of training: PartlyPaid gateway models carry a zero-retention, no-training commitment, but several free and stealth models explicitly may use your data to improve the model.Our providers follow a zero-retention policy and do not use your data for model training, with the following exceptions:opencode.ai/v2/docs/console/models
Data residency: PartlyThe gateway names US and EU hosting in the current docs while the Zen marketing page still says US only; running against your own provider puts residency in your hands.All our models are hosted in the US and EU.opencode.ai/v2/docs/console/models
Getting out: PartlyUsage exports to CSV and refunds are possible within 14 days on unused credits, but any balance left when a workspace is deleted is lost outright.Any remaining credit balance is forfeited.opencode.ai/v2/docs/console/workspaces
Certifications: PartlyA SOC 2 Type 2 report and security policies are offered through a trust centre under NDA; no ISO 27001 or GDPR certification is claimed on the site.Our Trust Center has everything: SOC 2 Type 2 report, security policies, subprocessor list, and answers to common security questions. Visit…opencode.ai/enterprise

Plans and pricing

Price and terms

Free plan: YesYesThe agent itself is free and open source with no account required, and ships with a set of free models
The terms in full
The free models are described as free "for a limited time" and several may use submitted data to improve the model; paid models are pay-as-you-go per 1M tokens, and a $10/month Go plan is the cheapest subscription.

Checked 21 October 2026.

Works with

Claude
GPT
Yes
Open models

For AI agents

MCP server: Partly
API: Yes

Developer docs

AdYour company across The Agents IndexAdvertise here

Screenshots

Docs, opencode.ai, captured 18 October 2026

Filed under

Coding agents Terminal coding agentsAPIAutonomousClaudeFree tierGPTMulti-agentOpen modelsOpen sourceSelf-hosted

Sources and updates

Last changed 21 October 2026

Sources39

Repository30 August 2026

Licence: MIT
Stars: 202,501
Forks: 26,324
Last push: 30 August 2026

The Agents Index in AI answers0 of 1

PerplexityDoes not cite The Agents IndexAsked OpenCode alternatives on 13 October 2026Other sources: alternativeto.net, openalternative.co, theaiagentindex.com, replicas.dev, zbuild.io, respan.ai

Updates6

21 October 2026Capabilities · Free tier · Terms
19 October 2026Capabilities · Free tier · Terms
21 September 2026Capabilities · Free tier · Price · Terms
2 September 2026Capabilities · Free tier · Terms
31 August 2026Capabilities · Free tier · Price · Terms
22 August 2026Alternatives · Repository · Summary · Tagline

Is this yours? Claim it free and edit it in your dashboard, or feature it in Coding agents.