openhuman
Agent Framework
The fastest, cheapest, most efficient open-source agent harness. Run more than 500 agents on a $10 VPS.
openhuman is an open-source agent framework AI agent maintained by tinyhumansai. It ranks #60 on The Agentic Leaderboard with an overall score of 94.1 / 100, including 97.4 / 100 on repository health (CI, tests, community health) and 100 / 100 on docs & examples (a docs folder, examples). The project has 42,712 GitHub stars.
Scores are estimated from public repository signals: CI, tests, documentation, maintenance and adoption. We do not run the agents, and small differences in rank are not meaningful. How scoring works
Add this badge to your README
[](https://www.theagenticleaderboard.com/agent/openhuman)
Rank history
scoring method changed this week20 weeks on the board since Jul 6 · best rank #60
Dashed line: the scoring method changed on Nov 16. Ranks before and after are not directly comparable. What changed
Compared to other Agent Framework agents
- #62 claude-mem— score 93.9
- #63 zeroclaw— score 93.9
- #64 composio— score 93.8