OpenHands
Self-hosted control center for AI coding agents that write code, review PRs and run automations.
Quick answer
OpenHands is a self-hosted control center for AI coding agents that write code, review PRs and run automations, listed under Coding assistants. A typical small-business use: a shared agent server that handles code review and dependency updates for the dev team. Official hardware: recommended 2 vCPU · 4 GB RAM (no separate minimum published). That fits a Hetzner Cloud CPX22 server at about €19.99/month (server only, excl. VAT). License: MIT, which allows internal business use.
Last updated · facts verified against official docs on · sources
Activity data comes from GitHub and was last refreshed Oct 22, 2026.
What is OpenHands?
OpenHands Agent Canvas is a web control center for software agents. It runs the open-source OpenHands agent, or third-party agents such as Claude Code, Codex, Gemini CLI and OpenCode, on your own machine, in Docker containers or on a VM. Agents can edit code, run commands and be triggered by schedules or webhooks from Slack, GitHub or Linear. You bring your own LLM.
Where it fits in a 10–200 person business
- A shared agent server that handles code review and dependency updates for the dev team
- Turn GitHub issues into tasks or draft fixes automatically
- Scheduled reports or maintenance jobs that post results to Slack
These use cases are our editorial suggestions, not claims made by the project.
Can a business use OpenHands? (license)
| License | MIT |
|---|---|
| Type | Permissive: MIT / Apache-2.0: use, modify and sell with few obligations. |
| Internal business use | Allowed under the license text |
| What to watch | Permissive. Internal and commercial use allowed; keep the license notice. LLM usage is billed by your model provider. |
This is a plain-English summary of the license text, not legal advice. Third-party models you run with it have their own licenses.
Data & compliance: does OpenHands phone home?
| Telemetry (usage data sent to the vendor) | Not documented Product analytics (PostHog) wait for the user's consent in the local UI. The example env file says VITE_DO_NOT_TRACK=1 disables all telemetry "including anonymous install counts". Source: github.com/OpenHands/OpenHands/blob/main/.env.sample |
|---|---|
| Other calls home by default (updates, marketplaces, default cloud services) | Not documented |
| Single sign-on (OIDC / SAML / LDAP) | Not documented |
| Audit log | Not documented |
| Role-based access control (RBAC) | Not documented |
| Data residency / GDPR notes from the project | Not documented |
Checked against the project's own docs and repository on . "Not documented" means we could not find the answer in official sources; it does not mean the feature is missing or that no data is sent. Third-party services you connect (model APIs, vector databases, search engines) have their own data policies. Compare all apps.
What hardware does OpenHands need?
| Minimum | Not published |
|---|---|
| Recommended | 2 vCPU · 4 GB RAM |
The self-hosting guide says 2 vCPU / 4 GB RAM "is plenty for a single user" on a VM (Ubuntu 24.04 suggested). No figure is given for teams.
Source: github.com/OpenHands/OpenHands/blob/main/docs/SELF_HOSTING.md
How much does it cost to host OpenHands per month?
| Recommended spec | 2 vCPU · 4 GB RAM → CPX22 (2 vCPU, 4 GB, 80 GB disk) €19.99/mo |
|---|
How we got this: we took the cheapest Hetzner Cloud Regular Performance (shared AMD vCPU) plan with at least the RAM and vCPU the project lists, using prices read from www.hetzner.com/cloud/regular-performance on 2026-10-22. "From" monthly price cap shown by Hetzner, excluding VAT. Prices vary by location. Covers the server only; it excludes LLM API usage, backups, extra storage, GPUs and your time. Disk is not matched; add a volume if you need more than the plan includes.
How to deploy OpenHands
Method: Docker (or npm package)
export PROJECTS_PATH="$HOME/projects"
mkdir -p "$PROJECTS_PATH" "$HOME/.openhands"
docker run -it --rm \
-p 127.0.0.1:8000:8000 \
-e AGENT_CANVAS_ALLOW_LAN_SESSION_KEY=true \
-v "$HOME/.openhands:/home/openhands/.openhands" \
-v "${PROJECTS_PATH}:/projects" \
ghcr.io/openhands/agent-canvas:1.30.1
# then open http://localhost:8000/canvas From the official instructions: github.com/OpenHands/OpenHands#option-2-with-a-docker-sandbox. Always check the source for the current version before you run it.
Known gotchas when self-hosting OpenHands
- Anyone who can reach the agent server can read and write files, run shell commands and use the network on that machine. Firewall everything except SSH, set a strong LOCAL_BACKEND_API_KEY and use --public mode before exposing it. source
- Running without the Docker sandbox (npm install / agent-canvas) gives the agent full access to the host filesystem. source
- If you publish the Docker port on a LAN or public interface, drop AGENT_CANVAS_ALLOW_LAN_SESSION_KEY so the session key is not injected into the page, and enter the API key in the UI instead. source
- Agent Canvas is labelled beta in the README (OpenHands incubator program), so check the current README before following older tutorials. source
Alternatives in Coding assistants
- Tabby Self-hosted AI coding assistant (Copilot alternative) for your team. Compare OpenHands vs Tabby
OpenHands FAQ
How much RAM does OpenHands need?
The recommended spec is 2 vCPU · 4 GB RAM. The self-hosting guide says 2 vCPU / 4 GB RAM "is plenty for a single user" on a VM (Ubuntu 24.04 suggested). No figure is given for teams.
How much does it cost to self-host OpenHands?
The software is free to download. Based on the official recommended 2 vCPU · 4 GB RAM (no separate minimum published): That fits a Hetzner Cloud CPX22 server at about €19.99/month (server only, excl. VAT). That covers the server only, not LLM API usage, backups, extra storage or your time.
Is OpenHands free for commercial use?
OpenHands is licensed under MIT (Permissive). Internal business use is allowed under the license text. Permissive. Internal and commercial use allowed; keep the license notice. LLM usage is billed by your model provider. This is a summary, not legal advice.
What are common problems when self-hosting OpenHands?
Anyone who can reach the agent server can read and write files, run shell commands and use the network on that machine. Firewall everything except SSH, set a strong LOCAL_BACKEND_API_KEY and use --public mode before exposing it. Running without the Docker sandbox (npm install / agent-canvas) gives the agent full access to the host filesystem. If you publish the Docker port on a LAN or public interface, drop AGENT_CANVAS_ALLOW_LAN_SESSION_KEY so the session key is not injected into the page, and enter the API key in the UI instead. Agent Canvas is labelled beta in the README (OpenHands incubator program), so check the current README before following older tutorials.
Is OpenHands still maintained?
Yes, it is actively developed. Last commit: Oct 22, 2026. Latest release: v1.30.1 (Oct 22, 2026). Checked Oct 22, 2026.
How do I install OpenHands?
The official method is Docker (or npm package). The exact commands are in the "How to deploy" section above and in the official instructions: https://github.com/OpenHands/OpenHands#option-2-with-a-docker-sandbox
What are self-hosted alternatives to OpenHands?
Other apps in Coding assistants: Tabby.
Sources
- GitHub repository: github.com/OpenHands/OpenHands
- Official documentation, Hardware requirements, Gotcha 1: github.com/OpenHands/OpenHands/blob/main/docs/SELF_HOSTING.md
- License text: github.com/OpenHands/OpenHands/blob/main/LICENSE
- Deploy instructions: github.com/OpenHands/OpenHands#option-2-with-a-docker-sandbox
- Gotcha 2: github.com/OpenHands/OpenHands#option-1-without-a-sandbox
- Gotcha 3: github.com/OpenHands/OpenHands#quickstart
- Gotcha 4: github.com/OpenHands/OpenHands#readme
- Hetzner Cloud pricing (cost estimate): www.hetzner.com/cloud/regular-performance
- GitHub stars, commits and releases: GitHub REST API, fetched
Facts on this page were checked against the project's own repository and docs on 2026-10-22. Spotted something out of date? Tell us.