OpenHands · Tools

OpenHands

Visit website

At a glance

Price
Pricing on request
Vendor
OpenHands

Specifications & properties

Key decision factors

Pricing model
Open source 1
Tool form
  • CLI agent
  • Web app
1
Autonomy level
Autonomous agent 1
Model choice
Bring your own key (BYO) 1

Pricing

Price from
Pricing on request 1
Free tier
Yes 1

Integration

API available
Yes 1

Compliance

Self-hosting / on-prem
Yes 1
Report data / suggest a correction

OpenHands is an MIT-licensed open-source AI software development agent that runs locally, in the cloud, or self-hosted in your VPC.

Profile

OpenHands is an open-source AI agent platform, formerly known as OpenDevin, that runs autonomous software-engineering agents rather than just suggesting completions: it plans, writes, and applies changes across a codebase, runs terminal commands, browses the web, and opens pull requests inside a sandboxed Docker environment. It is built by OpenHands, and the core project is free to self-host under the MIT license.

Who builds it

  • Developed as an open-source project at github.com/OpenHands/OpenHands, with 76,000+ GitHub stars and 9,700+ forks reported as of mid-2026.
  • The project is model-agnostic, so it can be pointed at different LLM providers rather than being locked to one vendor.
  • Customer-reported adoption includes teams at TikTok, Amazon, Netflix, and Google, per OpenHands' own site.

Core features

  • Agent Canvas: a workspace for managing and monitoring multiple agents at once.
  • Cloud execution: agents can keep running in isolated cloud sandboxes even when a developer's own machine is off.
  • Vulnerability management: scans repositories for security issues and proposes fixes as reviewable pull requests.
  • PR review automation: quality and security assessment on pull requests.
  • Code migration: assists with modernizing legacy systems, for example moving COBOL to Java.
  • Incident triage: investigates errors and helps identify root causes.
  • Integrates with GitHub, GitLab, CI/CD pipelines, Slack, and ticketing tools; an SDK is available for custom builds on top of the agent.

Pricing

  • Individual / Open Source: the local, self-hosted agent is free, including the web GUI, terminal UI, CLI, and Git integrations, with community support.
  • Individual (cloud/SaaS): also free, adding hosted cloud access, Jira and Slack integrations, and bring-your-own-key or pay-at-cost LLM access, but capped at 10 conversations per day per user.
  • Enterprise: custom pricing for SaaS or fully self-hosted VPC deployment, with SSO/SAML, unlimited concurrent conversations, a large-codebase SDK, and priority support with a named customer engineer — available only after contacting the OpenHands team.

Who it's for

OpenHands suits teams that want a fully open, self-hostable autonomous coding agent rather than a closed SaaS product, whether that means running it locally for individual use, adopting the free cloud tier for light day-to-day use, or moving to Enterprise for VPC/self-hosted deployment, SSO, and unlimited usage at scale. Its focus on end-to-end task execution — not just suggestions — also makes it a fit for teams specifically looking to automate vulnerability fixes, PR review, and legacy code migration rather than just in-editor autocomplete.

Frequently asked questions

Is OpenHands free to use?

Yes — the core OpenHands agent is open source and free to self-host, and its hosted Individual/SaaS tier is also free, though capped at 10 conversations per day per user. Users can bring their own LLM API key or pay OpenHands' provider at cost with no markup.

What does OpenHands cost for teams or enterprises?

Enterprise pricing is custom and only available after contacting the OpenHands team; it covers SaaS or fully self-hosted VPC deployment with SSO/SAML, unlimited concurrent conversations, and priority support with a named customer engineer, on top of everything in the free tiers.

Is OpenHands open source and can it be self-hosted?

Yes — OpenHands is released under the MIT license with its source at github.com/OpenHands/OpenHands, and it can be run entirely locally or self-hosted in a private VPC as part of the Enterprise plan, keeping code and data inside a customer's own infrastructure.

What can OpenHands agents actually do beyond code suggestions?

OpenHands agents plan and execute full engineering tasks: they write and edit code across a codebase, run terminal commands, browse the web, and open pull requests, all inside a sandboxed Docker environment rather than just proposing inline completions.

Does OpenHands work with any AI model, or only specific ones?

OpenHands is model-agnostic, meaning it can be configured to work with a range of LLM providers rather than being locked into a single vendor's models, giving teams flexibility to choose or swap the underlying model.

Does OpenHands help with security, not just feature work?

Yes — OpenHands includes vulnerability management that scans repositories for security issues and proposes fixes as reviewable pull requests, plus automated PR review for quality and security assessment, alongside its code-writing and migration capabilities.