OpenHands
Last tested:
OpenHands is open-source platform for software engineering agents (MIT). Self-host with Agent Canvas, a headless CLI, an SDK, or use OpenHands Cloud. Bring your own LLM key; Cloud also offers at-cost hosted models and a limited free BYOK tier (category: Coding Agents (CLI & Cloud)). VibeToolRank agent test score: 9.1/10 (tested 2026-11-04, 3 runs, headless CLI, BYOK), T-6 of 15 tested Coding Agents (CLI & Cloud) tools. Pricing: Agent Canvas free to self-host (MIT, BYOK, single shared identity). Teams $500/mo up to 10 users (+$50/seat), self-hosted with RBAC. Enterprise custom. OpenHands Cloud: free BYOK tier with daily/parallel conversation limits; hosted models at provider cost with no markup. Free tier: Yes: self-host Agent Canvas / CLI / SDK free with your own model key; Cloud free BYOK tier for individuals with usage limits. No card required for local BYOK.
Key facts
Verdict: OpenHands CLI headless fixed our seeded cart discount bug in 3 of 3 runs (42–51s each) with a free Gemini key, never touched the test file.
- VibeToolRank score
- 9.1/10
- Rank
- T-6 of 15 tested Coding Agents (CLI & Cloud)
- Category
- Coding Agents (CLI & Cloud)
- Last tested
- 2026-11-04
- Runs
- 3 runs
- Outcome score
- 10/10
- Edition tested
- OpenHands CLI 1.16.0 headless with Google AI Studio gemini-3.5-flash-lite and caching_prompt=False (BYOK, $0). Not Agent Canvas, not Cloud
- Free plan, as measured
- $0 per bugfix run on a free Gemini key (BYOK)
- Free plan
- Yes: self-host Agent Canvas / CLI / SDK free with your own model key; Cloud free BYOK tier for individuals with usage limits. No card required for local BYOK
- Price
- Agent Canvas free to self-host (MIT, BYOK, single shared identity). Teams $500/mo up to 10 users (+$50/seat), self-hosted with RBAC. Enterprise custom. OpenHands Cloud: free BYOK tier with daily/parallel conversation limits; hosted models at provider cost with no markup
- Pricing checked
- 2026-11-04
- Sources
- openhands.dev/pricing · openhands.dev · docs.openhands.dev/overview/introduction · test log in our repo
Pricing and free tier
Current official pricing (checked ): Agent Canvas free to self-host (MIT, BYOK, single shared identity). Teams $500/mo up to 10 users (+$50/seat), self-hosted with RBAC. Enterprise custom. OpenHands Cloud: free BYOK tier with daily/parallel conversation limits; hosted models at provider cost with no markup
Is OpenHands free? Yes: self-host Agent Canvas / CLI / SDK free with your own model key; Cloud free BYOK tier for individuals with usage limits. No card required for local BYOK
Local self-host is free; you pay only your model provider. Gemini free key works once caching_prompt is off. OpenRouter :free models are rate-limited (~50 requests/day).
verified-official official source · full-page-read. Official plan limits can change; re-check the vendor page before paying. Figures from our tests are recorded observations, not a current-capacity promise.
What we measured in our test (2026-11-04, 3 runs): $0 per bugfix run on a free Gemini key (BYOK) OpenHands free tier: how far you get
Agent test verdict
Edition tested OpenHands CLI 1.16.0 headless with Google AI Studio gemini-3.5-flash-lite and caching_prompt=False (BYOK, $0). Not Agent Canvas, not Cloud. Results may differ for the hosted product.
Tested 2026-11-04 · 3 runs · rubric v0.1 (provisional thresholds). OpenHands CLI headless fixed our seeded cart discount bug in 3 of 3 runs (42–51s each) with a free Gemini key, never touched the test file. Cost $0. Overall 9.1 (outcome 10, speed 8, cost 10, DX 6, reliability 10). DX lost points because Gemini free context caching caused 429s until we seeded agent_settings and forced caching_prompt=False — there is no simple CLI/env switch.
| Dimension | Score /10 |
|---|---|
| Outcome (40%) | 10 |
| Speed (15%) | 8 |
| Cost (15%) | 10 |
| Developer experience (15%) | 6 |
| Reliability (15%) | 10 |
| Overall | 9.1 |
How we tested
- OpenHands CLI 1.16.0 in a scratch uv venv, headless, bubblewrap sandbox.
- Model: Google AI Studio gemini-3.5-flash-lite with caching_prompt forced false in agent_settings.
- Task T3: fix seeded cart.js so all 4 node --test tests pass without editing the test file.
- 3 runs, all passed (4/4) in 41.6–50.6s, $0.
Pros, cons and who it's for
Pros
- Fixed the seeded bug in 3/3 headless runs with a free model
- Did not modify the test file in any run
- Self-host / BYOK path works with no OpenHands Cloud account
- Large open-source project (~92.6k GitHub stars on 2026-11-04)
Cons
- Default Gemini path uses context caching that the free Gemini tier rejects (429); turning it off needed a settings seed + JSON patch
- CLI TUI is in maintenance mode; docs steer new local users to Agent Canvas
- We only scored the headless CLI edition, not Agent Canvas or Cloud
Who it's for: Teams that want an open-source coding agent they can self-host or run headless in CI with their own model keys, without locking into one cloud vendor.
Free-tier limits: Local self-host is free; you pay only your model provider. Gemini free key works once caching_prompt is off. OpenRouter :free models are rate-limited (~50 requests/day).
FAQ
Is OpenHands free?
Yes: self-host Agent Canvas / CLI / SDK free with your own model key; Cloud free BYOK tier for individuals with usage limits. No card required for local BYOK. In our test: $0 per bugfix run on a free Gemini key (BYOK).
How did OpenHands score in VibeToolRank's test?
9.1/10 overall (outcome 10, speed 8, cost 10, developer experience 6, reliability 10), 3 runs on 2026-11-04. OpenHands CLI headless fixed our seeded cart discount bug in 3 of 3 runs (42–51s each) with a free Gemini key, never touched the test file.
How much does OpenHands cost?
Agent Canvas free to self-host (MIT, BYOK, single shared identity). Teams $500/mo up to 10 users (+$50/seat), self-hosted with RBAC. Enterprise custom. OpenHands Cloud: free BYOK tier with daily/parallel conversation limits; hosted models at provider cost with no markup (from the official pricing page, checked 2026-11-04).
Related
Comparisons
Other Coding Agents (CLI & Cloud)
- Aider — 9.3/10
- Claude Code — 9.3/10
- Cline — 9.2/10
- GitHub Copilot CLI — 9.2/10
- OpenCode — 9.2/10
- Memdoor — 9.1/10
- Zerg Client — 9.1/10
- Amp — 9/10
- Reason — 8.6/10
- Alysis Code — 8.4/10
- Harness (Autonomous) — 8.4/10
- Jules — 8/10
- OpenAI Codex — 7.4/10
- Gemini CLI — 6.1/10
- Agent Manager — not yet tested
- Conducted — not yet tested
- Crewplane — not yet tested
- Devin — not yet tested
- Mirror (Reflection) — not yet tested
- Parallax (Conifer) — not yet tested
- psi — not yet tested
- Roomote (formerly Roo Code) — not yet tested
- Ski — not yet tested
- Wonder — not yet tested
Best Coding Agents (CLI & Cloud) · Best free Coding Agents (CLI & Cloud) · All tools · How we test
Sources
Scores come from our own tests and are never for sale; sponsored placements are always labelled. Some links are affiliate links (currently Railway). Disclosure