Back to directory

Langroid

A lightweight Python framework for building LLM-powered applications by setting up multiple agents, equipping them with models, vector stores and tools, and having them collaboratively solve tasks by exchanging messages.

Langroid

Langroid is a Python framework for building LLM-powered applications around a multi-agent paradigm inspired by the Actor Framework. Developers set up agents, optionally equip them with a language model, a vector store and tools or functions, assign them tasks, and let them collaboratively solve a problem by exchanging messages.

It does not depend on LangChain or another LLM framework, and works with practically any LLM. It includes an adapter that lets an agent use Model Context Protocol servers by converting their tools into Langroid's own tool message format, and is used in production for tasks such as multi-agent orchestration for vulnerability detection in secure software development.

It was created by researchers from Carnegie Mellon University and the University of Wisconsin-Madison, and is licensed under MIT.

More in Agent frameworks

See all 902
  • Recurse

    recurse.run

    A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.

  • ProofForge

    shulgin.is-a.dev

    An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.

  • local-coder

    github.com

    Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.