{"slug":"openhuman","name":"OpenHuman","vendor":"Tiny Humans","tagline":"Local-first personal agent harness with a memory tree, agent fleets and 17 messaging channels","url":"https://agentboards.org/agent/openhuman","website":"https://tinyhumans.ai/openhuman","docs":"https://tinyhumans.gitbook.io/openhuman/","repo":"https://github.com/tinyhumansai/openhuman","category":"harness","execution":["local"],"license":"GPL-3.0","open_source":true,"pricing":{"model":"mixed","summary":"Open source under GPL-3.0 and free to run with your own provider key or a local Ollama model; an OpenHuman subscription bundles managed inference, Exa web search and media generation","from_usd":null,"free_tier":true,"byok":true},"models":{"backbone":["any"],"bring_your_own_model":true,"local_models":true},"protocols":{"mcp_client":true,"mcp_server":false},"capabilities":{"terminal_exec":true,"browser":true,"multi_file_edit":false,"git_ops":false,"docker_sandbox":false,"multi_agent":true,"headless_ci":false},"context_window":null,"platforms":["macos","linux","windows"],"install":[],"benchmarks":[],"tags":["personal-agent","memory","orchestrator","local-first","preview"],"github_stars":40434,"metrics":{"stars":40434,"pushed_at":"2026-10-01T19:48:14Z","archived":false,"open_issues":255,"latest_version":"v0.64.10","latest_version_at":"2026-09-30T21:18:38Z","latest_version_url":"https://github.com/tinyhumansai/openhuman/releases/tag/v0.64.10","version_source":"github","checked_at":"2026-10-02T05:02:02.303Z"},"sources":{"overview":"https://github.com/tinyhumansai/openhuman","docs":"https://tinyhumans.gitbook.io/openhuman/"},"verified_at":null,"adoption":{"score":7.1,"signals":{"github_stars":39904,"hn_mentions_1y":4},"measured_at":"2026-09-21"},"scores":{"panel":5.5,"adoption":7.1,"spread":4,"facts":0,"board":4.9,"dimensions":{"reliability":4.7,"usefulness":5.5,"cost":7.3,"longevity":4.5},"per_persona":{"amigo":5.5,"critico":4.8,"profesor":6,"inversora":5.3,"jefa":3.8,"hacker":7.8},"community":null},"rank":null,"category_rank":175,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"Four points between El Hacker's GPL fork and La Jefa's data protection file, and the word neither of them disputes is preview.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick OpenHuman if you want an agent that reaches you in the apps you already use all day; pick Letta if you want the memory without also inviting it into your messages.","scores":{"reliability":4,"usefulness":6,"cost":8,"longevity":4},"evidence":["https://tinyhumans.ai/openhuman","https://github.com/tinyhumansai/openhuman"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"A preview-stage personal agent with command execution and a browser, accumulating a database of your life, with no isolation layer named anywhere.","scores":{"reliability":3,"usefulness":5,"cost":7,"longevity":4},"evidence":["https://github.com/tinyhumansai/openhuman","https://tinyhumans.gitbook.io/openhuman/"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"Memory is compressed into Markdown trees held in SQLite, and sub-agent fleets execute on checkpointed graphs, so both retention and resumption are explicit mechanisms rather than prompt tricks.","scores":{"reliability":6,"usefulness":6,"cost":7,"longevity":5},"evidence":["https://tinyhumans.gitbook.io/openhuman/","https://github.com/tinyhumansai/openhuman"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"The subscription bundles managed inference, web search and media generation, which means the business is reselling three upstream vendors at a markup you cannot see.","scores":{"reliability":5,"usefulness":6,"cost":6,"longevity":4},"evidence":["https://tinyhumans.ai/openhuman","https://github.com/tinyhumansai/openhuman"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"This reads employee communications across seventeen personal channels, which is a data protection question before it is a tooling question, and the answer is no.","scores":{"reliability":2,"usefulness":3,"cost":7,"longevity":3},"evidence":["https://tinyhumans.ai/openhuman","https://tinyhumans.gitbook.io/openhuman/"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"GPL-3.0 means a fork stays open, it takes a local Ollama endpoint, it speaks MCP as a client, and there is no published install command, so I clone it anyway.","scores":{"reliability":8,"usefulness":7,"cost":9,"longevity":7},"evidence":["https://github.com/tinyhumansai/openhuman","https://tinyhumans.gitbook.io/openhuman/"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"Seventeen messaging channels, which is sixteen additional ways to be told the task failed.","scores":null,"evidence":[]}]}