OpenHands

Open source AI agent platform for cloud coding agents, from free local use to enterprise self-hosted deployment.

Best forOpen source-minded engineering teams and enterprises

Visit OpenHands Compare

Our verdict

OpenHands is a credible, community-backed choice for engineering teams that want an open source, self-hostable coding agent platform rather than a closed SaaS tool.

Several of the strongest claims on this listing, including the 87% same-day bug-fix rate, the $18.8M funding figure, and the 81K/80.8K star counts, come from a customer testimonial and a quoted tweet on OpenHands' own marketing page rather than independently audited sources, so treat them as vendor-reported figures.

The Agents Index digest

New agents, fresh verdicts, and who's earning the top spots. One short email.

Screenshot of OpenHands
OpenHands, captured at verification, . Click to enlarge.

What's great

  • Fully free and open source core, MIT licensed, for local single-user use.
  • OpenHands SDK lets teams embed the agent harness into their own products or internal platforms.
  • Large Codebase SDK is purpose-built for large or legacy codebases with parallel multi-agent orchestration.
  • Enterprise self-hosted or private-VPC deployment with SSO, RBAC, and audit logs for security-conscious organizations.
  • Broad integration surface spanning GitHub, GitLab, CI/CD, Slack, Linear, and 70+ other tools.

Watch-outs

  • Even the built-in at-cost model option is pay-as-you-go rather than bundled into the flat subscription, so ongoing model usage costs apply whether you bring your own key or use OpenHands' own provider.
  • A third-party review flags complex installation and setup that requires technical skill, making it less approachable for non-technical users.
  • The free Individual Cloud tier is capped at 1 user and 10 daily conversations, limiting it for solo, light-usage scenarios.
  • A third-party composite evaluation scores OpenHands 68/100 and ranks it #19 of 27 in the AI Agent (Coding) category, suggesting mixed standing versus competitors on that site.
  • The MIT license covers everything outside the repository's enterprise/ directory; that directory's code is under a separate PolyForm Free Trial License (source-available, 30-day trial only without a commercial license) — not the full-repo MIT grant the "Open Source" plan card implies.

Also note: Its hosted model option is billed at cost on a pay-as-you-go basis, not included free in a flat subscription. · Individual Cloud plan is limited to 1 user and a maximum of 10 daily conversations. · Reported context length is 128K tokens. · Interface and documentation are English only.

What is OpenHands?

OpenHands (formerly OpenDevin) is an open source AI agent platform for software development, built by All-Hands-AI as the open platform for cloud coding agents. It serves individual developers, engineering teams, and enterprises through a free local OSS release, a hosted Cloud tier, and self-hosted enterprise deployment. With over 81,000 GitHub stars, OpenHands sits within the broader open source autonomous coding agent ecosystem, appealing to technical teams that want agents capable of reading, editing, and executing code across codebases while retaining control over deployment and data.

What does OpenHands do?

OpenHands functions as an agent harness that lets AI models plan, write, edit, and execute code against real repositories rather than just suggesting snippets in a chat window. Developers run OpenHands locally for free with their own LLM API key (such as Claude or GPT), or use the hosted Individual Cloud tier, which lets you bring that same API key or instead use OpenHands' own models on a pay-as-you-go basis, billed at cost with no markup. The platform's Large Codebase SDK maps dependencies across large or legacy codebases and orchestrates changes so multiple agents can work in parallel without stepping on each other's edits. A separate OpenHands SDK lets engineering teams embed the same agent harness into their own products or internal developer platforms rather than using OpenHands only as a standalone tool. Individual users get local installation, parallel agents on their own machine, custom automations, an Agent Canvas backed by a VM, plus the Cloud tier's hosted access from desktop and mobile, API support, and Jira/Slack integrations. Enterprise deployments add SSO, RBAC, audit logs, budget controls, and sandboxed execution inside a customer's own VPC or fully self-hosted environment, so code and data never leave the organization's control. Native integrations connect OpenHands to GitHub, GitLab, CI/CD pipelines, Slack, and ticketing systems, with automation templates extending to 70+ third-party tools including Linear.

Key features

  • Large Codebase SDK: Maps dependencies across large or legacy codebases and orchestrates changes so multiple agents can work in parallel without conflicting edits.
  • OpenHands SDK: An embeddable agent harness that lets teams build agentic AI directly into their own products or internal developer platforms.
  • Agent Canvas with VM backend: Included in the free Individual tier for running and building automations locally, with parallel agents on your own machine.
  • 70+ native integrations: Connects natively to GitHub, GitLab, CI/CD pipelines, Slack, and ticketing tools, with automation templates extending to Linear and other services.
  • Self-hosted, sandboxed execution: Enterprise deployments run in a customer's own VPC or fully self-hosted environment with SSO, RBAC, audit logs, and budget controls so code never leaves the organization's control.

What are OpenHands's use cases?

  • Local bug fixing with your own API key: An individual developer installs OpenHands for free, connects a Claude or GPT API key, and runs parallel agents locally to triage and fix bugs.
  • Team-wide coding automations: An engineering team uses the Team plan to share automations, invite org members, and let cloud-hosted agents respond to GitHub and Slack activity even when the team is offline.
  • Large or legacy codebase modernization: An organization uses the Large Codebase SDK to map dependencies and run multiple agents in parallel across a sprawling legacy repository without conflicting changes.
  • Enterprise-grade secure deployment: A security-conscious enterprise self-hosts OpenHands in its own VPC with SSO, RBAC, and audit logs so agent-driven code changes stay inside its compliance boundary.

What does OpenHands integrate with?

  • GitHub
  • GitLab
  • CI/CD pipelines
  • Slack
  • Linear
  • 70+ automation-template integrations
  • Model Context Protocol (MCP) — connects as a client to external MCP servers via SSE, Streamable HTTP or Stdio

Why use OpenHands?

  • 81K+ GitHub stars reflect substantial open source community adoption and active development.
  • OpenHands reports SWE-Bench verified performance as a benchmark of real-world code-fixing capability.
  • A customer testimonial cited on the site reports OpenHands autonomously fixes 87% of bug tickets same-day.
  • The project is backed by $18.8M in funding, per a tweet quoted on the OpenHands site, to build out cloud coding agents.

Pros & cons

Pros

  • Fully free and open source core, MIT licensed, for local single-user use.
  • OpenHands SDK lets teams embed the agent harness into their own products or internal platforms.
  • Large Codebase SDK is purpose-built for large or legacy codebases with parallel multi-agent orchestration.
  • Enterprise self-hosted or private-VPC deployment with SSO, RBAC, and audit logs for security-conscious organizations.
  • Broad integration surface spanning GitHub, GitLab, CI/CD, Slack, Linear, and 70+ other tools.

Cons

  • Even the built-in at-cost model option is pay-as-you-go rather than bundled into the flat subscription, so ongoing model usage costs apply whether you bring your own key or use OpenHands' own provider.
  • A third-party review flags complex installation and setup that requires technical skill, making it less approachable for non-technical users.
  • The free Individual Cloud tier is capped at 1 user and 10 daily conversations, limiting it for solo, light-usage scenarios.
  • A third-party composite evaluation scores OpenHands 68/100 and ranks it #19 of 27 in the AI Agent (Coding) category, suggesting mixed standing versus competitors on that site.
  • The MIT license covers everything outside the repository's enterprise/ directory; that directory's code is under a separate PolyForm Free Trial License (source-available, 30-day trial only without a commercial license) — not the full-repo MIT grant the "Open Source" plan card implies.

Limitations

  • Its hosted model option is billed at cost on a pay-as-you-go basis, not included free in a flat subscription.
  • Individual Cloud plan is limited to 1 user and a maximum of 10 daily conversations.
  • Reported context length is 128K tokens.
  • Interface and documentation are English only.

OpenHands pricing

Free open source and free Individual Cloud tier (10 conversations/day, 1 user); Enterprise is custom-priced.

  • Open SourceFree
  • Individual (Cloud)$0 / /mo
  • EnterpriseCustom pricing / contact sales

See current pricing on all-hands.dev ↗Compare OpenHands alternatives →

OpenHands specs

Pricing

Pricing model
freemium
Free tier
✓ Yes

Capabilities

Model / LLM
Bring your own API key, or use OpenHands' own models at cost with no markup — model-agnostic (commonly Claude or GPT)
Interface
Web
Public API
✓ Yes
Open source
✓ Yes

Deployment

Deployment
Both

OpenHands review

OpenHands is a credible, community-backed choice for engineering teams that want an open source, self-hostable coding agent platform rather than a closed SaaS tool. Its free OSS core, embeddable SDK, and enterprise-grade sandboxed deployment options make it flexible for both solo developers and security-conscious organizations, provided you're willing to accept a more technical setup process — whether you bring your own LLM API key or pay OpenHands' own at-cost rate for model usage.

Several of the strongest claims on this listing, including the 87% same-day bug-fix rate, the $18.8M funding figure, and the 81K/80.8K star counts, come from a customer testimonial and a quoted tweet on OpenHands' own marketing page rather than independently audited sources, so treat them as vendor-reported figures.

Frequently asked questions

Is OpenHands free to use?
Yes. The Open Source version is free and MIT licensed for a single local user, and the Individual Cloud tier is also free but capped at 10 daily conversations for one user (model usage bills separately, whether BYOK or OpenHands' own at-cost option). The custom-priced Enterprise plan adds self-hosted/private-VPC deployment and unlimited usage.
Is all of OpenHands MIT licensed, including Enterprise?
No. The MIT license in the repository's root LICENSE file covers everything outside the enterprise/ directory. Code under enterprise/ is licensed separately under the PolyForm Free Trial License 1.0.0 — a source-available license limited to a 30-day free trial per calendar year without a paid commercial license. The self-hostable core you'd run locally or in the Open Source/Individual tiers is unaffected; the carve-out only applies to the Enterprise-specific code.
Does OpenHands provide its own AI models or API access?
Optionally. The hosted Individual Cloud tier lets you bring your own LLM API key (such as Claude or GPT) or instead use OpenHands' own models on a pay-as-you-go basis, billed at cost with no markup. The OpenHands SDK provides an agent harness and automation API support either way.
Can OpenHands be self-hosted for security or compliance reasons?
Yes. The Enterprise plan supports self-hosted or private-VPC deployment where code and data never leave the organization's environment, along with SSO, RBAC, audit logs, and budget controls.
What is the Large Codebase SDK?
It's a component that maps dependencies across large or legacy codebases and orchestrates changes so multiple agents can work on different parts of a codebase in parallel without conflicting edits.
What integrations does OpenHands support?
OpenHands natively integrates with GitHub, GitLab, CI/CD pipelines, Slack, and ticketing tools, plus automation templates connecting to 70+ third-party tools including Linear.
How easy is OpenHands to set up?
A third-party review notes OpenHands has complex installation and setup that requires technical skill, so it suits developers comfortable configuring agent environments rather than non-technical users wanting a plug-and-play tool.

Get a quote from OpenHands

Tell us what you need. We make the introduction to OpenHands.

OpenHands

from Free

Visit ↗