OpenHands
Free tierThe Open Platform for Cloud Coding Agents
Free tier available·All audiences·Powered by Multi-model / bring your own LLM (supports Claude, Codex, Gemini CLI, Kimi, DeepSeek, Qwen, and others)·API available·Open source
Key strengths
Open source (MIT licensed) and free to self-hostMulti-agent and multi-model support (Claude, Codex, Gemini CLI, open weights)Agentic SDLC automation (PR review, CI fix, issue triage, security scanning)Flexible deployment (local, cloud, VPC)SDK for building custom agent workflowsCentralized Agent Control Plane for enterprise governance
Free tier + paid plans · from $499 USD/mo
Founded 2024
Self-hostable
Facts verified against openhands.dev on
Recent changes
- Starting price: $0/mo → $499/mo
- Plans updated: Free tier (open source, self-hosted, MIT); Teams at $499/mo for up to 10 users (+$50/mo per additional seat, capped at 5,000 sessions/month); Enterprise at custom pricing with unlimited seats and uncapped usage.
- Models updated: Multi-model / bring your own LLM (supports Claude, Codex, Gemini CLI, Kimi, DeepSeek, Qwen, and others)
Get alerts when OpenHands changes
Free. We'll email you when pricing, the free tier, API access or open-source status changes.
No ratings yet
Technical Use Cases
OpenHands is purpose-built for agentic automation across the software development lifecycle. Key developer and engineering team use cases include:
- Automated PR review: Agents analyze pull requests, flag issues, and suggest or apply fixes — integrated directly with GitHub workflows.
- CI failure remediation: Agents detect broken CI pipelines and autonomously attempt to diagnose and fix the underlying code or configuration issues.
- Issue triage: Connect to Jira or Linear; agents classify, prioritize, and route incoming issues without manual intervention.
- Security scanning: Agents run security checks as part of the SDLC, surfacing vulnerabilities early in the development process.
- Custom agent workflows (SDK): Use the OpenHands SDK to compose multi-agent pipelines for bespoke automation — e.g., chaining issue ingestion → code generation → test execution → PR creation.
- Multi-model experimentation: Swap or combine LLM backends (Claude, Codex, Gemini CLI, DeepSeek, Qwen, etc.) within the same workflow to benchmark or optimize agent performance.
- Enterprise governance: The Centralized Agent Control Plane provides audit trails, access controls, and usage visibility across all agents and teams.
- VPC / air-gapped deployment: Self-host within a private cloud environment for compliance-sensitive workloads.
Alternatives
See all alternatives