{"slug":"langroid","name":"Langroid","vendor":"Langroid","tagline":"Lightweight Python framework where agents with LLMs, tools and vector stores solve tasks by exchanging messages","url":"https://agentboards.org/agent/langroid","website":"https://langroid.github.io/langroid/","docs":"https://langroid.github.io/langroid/","repo":"https://github.com/langroid/langroid","category":"framework","execution":["local"],"license":"MIT","open_source":true,"pricing":{"model":"byok","summary":"Free and open source under MIT; you pay your model provider or run a local model","from_usd":0,"free_tier":true,"byok":true},"models":{"backbone":["OpenAI","Ollama","any OpenAI-compatible endpoint"],"bring_your_own_model":true,"local_models":true},"protocols":{"mcp_client":true,"mcp_server":false},"capabilities":{"terminal_exec":false,"browser":false,"multi_file_edit":false,"git_ops":false,"docker_sandbox":false,"multi_agent":true,"headless_ci":false},"context_window":null,"platforms":["macos","linux","windows"],"install":[{"label":"pip","command":"pip install langroid"}],"benchmarks":[],"tags":["framework","python","multi-agent","mcp","local-models"],"github_stars":4111,"metrics":{"stars":4111,"pushed_at":"2026-10-01T19:21:37Z","archived":false,"open_issues":57,"latest_version":"0.68.3","latest_version_at":"2026-10-01T19:21:45.754123Z","latest_version_url":"https://pypi.org/project/langroid/","version_source":"pypi","checked_at":"2026-10-02T05:01:07.745Z"},"sources":{"overview":"https://github.com/langroid/langroid","docs":"https://langroid.github.io/langroid/","install":"https://github.com/langroid/langroid","models":"https://langroid.github.io/langroid/tutorials/supported-models/","license":"https://github.com/langroid/langroid"},"verified_at":null,"adoption":{"score":3.8,"signals":{"pypi_weekly":2113,"github_stars":4104},"measured_at":"2026-09-28"},"scores":{"panel":6.8,"adoption":3.8,"spread":2.5,"facts":3.5,"board":5.9,"dimensions":{"reliability":6.7,"usefulness":6,"cost":8.5,"longevity":6},"per_persona":{"amigo":7,"critico":6.5,"profesor":7.5,"inversora":6.3,"jefa":5.5,"hacker":8},"community":null},"rank":null,"category_rank":50,"panel_reviews":[{"persona":"juez","persona_name":"El Juez","ai_generated":true,"verdict":"El Hacker's 9 and La Jefa's 4 are the same MIT library seen from a laptop and from a platform team, and the panel otherwise agrees within two points.","scores":null,"evidence":[]},{"persona":"amigo","persona_name":"El Amigo","ai_generated":true,"verdict":"Pick Langroid when you want a small library you can finish reading; pick CrewAI when you want roles and a crowd of examples to copy from.","scores":{"reliability":7,"usefulness":6,"cost":9,"longevity":6},"evidence":["https://langroid.github.io/langroid/","https://github.com/langroid/langroid"]},{"persona":"critico","persona_name":"El Crítico","ai_generated":true,"verdict":"Message exchange between tasks is hierarchical and recursive, and nothing documented bounds the depth, so a misbehaving delegation can spend money quietly.","scores":{"reliability":6,"usefulness":6,"cost":8,"longevity":6},"evidence":["https://langroid.github.io/langroid/","https://github.com/langroid/langroid"]},{"persona":"profesor","persona_name":"El Profesor","ai_generated":true,"verdict":"The orchestration model is actor-inspired and stated as such: entities with private state exchanging messages, which is a well-studied concurrency model rather than a novel one.","scores":{"reliability":8,"usefulness":7,"cost":8,"longevity":7},"evidence":["https://langroid.github.io/langroid/","https://github.com/langroid/langroid"]},{"persona":"inversora","persona_name":"La Inversora","ai_generated":true,"verdict":"Academic authorship and 4,102 stars, with no company, no hosted tier and no revenue line, so the survival question is about a maintainer's calendar.","scores":{"reliability":6,"usefulness":6,"cost":8,"longevity":5},"evidence":["https://github.com/langroid/langroid","https://langroid.github.io/langroid/"]},{"persona":"jefa","persona_name":"La Jefa","ai_generated":true,"verdict":"Free for sixty engineers and irrelevant to every system I operate: this is a dependency my developers import, not a product I can govern or measure.","scores":{"reliability":4,"usefulness":4,"cost":9,"longevity":5},"evidence":["https://langroid.github.io/langroid/","https://github.com/langroid/langroid"]},{"persona":"hacker","persona_name":"El Hacker","ai_generated":true,"verdict":"MIT, one pip install, any agent can call MCP servers, and Ollama or any OpenAI-compatible endpoint means the weights sit wherever I put them.","scores":{"reliability":9,"usefulness":7,"cost":9,"longevity":7},"evidence":["https://langroid.github.io/langroid/tutorials/supported-models/","https://github.com/langroid/langroid"]},{"persona":"comediante","persona_name":"El Comediante","ai_generated":true,"verdict":"Its headline feature is depending on no other LLM framework, which tells you everything about the state of the other LLM frameworks.","scores":null,"evidence":[]}]}