
langroid
Python framework where LLM agents wrap tasks and delegate work to sub-agents
- Pricing
- Open source
- Platforms
- self-hosted
- GitHub stars
- 4.2K
- Last commit
- 3 days ago
- Language
- Python
- License
- MIT
About langroid
langroid is an open-source agent framework for multi-agent frameworks. It is available on self-hosted.
Langroid treats an agent as the holder of conversation state, optional vector-store memory and tools. A task wraps an agent with instructions and runs it in turns across three responders: the LLM, the agent and the user. Tasks can contain sub-tasks, which behave as extra responders, giving hierarchical delegation.
The docs describe support for local and hosted models, Redis-based caching of prompts and responses, and the vector stores Qdrant, Chroma, LanceDB and Milvus for retrieval-augmented generation with source citations. The authors say the design draws on the Actor model. It suits Python developers who want a small, transparent orchestration layer.
What it does
- Agent and Task classes
- Agents hold state and tools; tasks give them instructions and manage turns.
- Hierarchical delegation
- Sub-tasks act as additional responders for recursive task splitting.
- Wide LLM support
- Works with local and remote models through several libraries.
- Vector-store retrieval
- Supports Qdrant, Chroma, LanceDB and Milvus for RAG.
- Prompt caching
- Uses Redis by default to cache prompts and responses.
Good for
- Build a multi-agent pipeline in Python
- Add retrieval over documents with source citations
- Delegate subtasks between specialised agents
- Run agents against local open models
Pricing
Open source
MIT licensed Python library
Prices change. Check langroid.github.io before you buy.
langroid FAQ
Which vector stores work?
Qdrant, Chroma, LanceDB and Milvus are listed as supported.
Can it use local models?
Yes, the docs say it works with local or remote models; guides cover non-OpenAI LLMs.
What licence applies?
MIT.
Is langroid free?
Yes. langroid is open source, so you can use and self-host it for free.
What are the best alternatives to langroid?
The closest alternatives to langroid are hermes-agent, deepseek-harness, langchain, deer-flow and ruflo.
Alternatives to langroid

hermes-agent
AgentOpen-source self-improving agent with memory, scheduling and many chat gateways
- Freemium
- Open-source agent
- 264K GitHub stars
- AI models and LLM tooling

deepseek-harness
AgentOpen-source agent harness from DeepSeek, built on a plugin architecture
- Open source
- Open-source agent
- 282K GitHub stars
- Multi-agent frameworks

langchain
AgentPython library for building agents from a model, tools, prompt and middleware
- Open source
- Agent framework
- 150K GitHub stars
- AI models and LLM tooling

deer-flow
AgentOpen-source super agent harness that researches, codes and creates with sandboxes
- Open source
- Open-source agent
- 85.4K GitHub stars
- AI models and LLM tooling

ruflo
AgentAgent meta-harness that adds swarms, memory and MCP tools to Claude Code and Codex
- Open source
- Agent framework
- 81.9K GitHub stars
- Knowledge and memory

autogen
AgentArchived Microsoft framework for multi-agent AI applications in Python
- Open source
- Agent framework
- 61.9K GitHub stars
- Multi-agent frameworks