Langroid
by ex-CMU/UW-Madison researchers
Multi-agent programming framework for LLM applications.
Agent Details
Quick Links
About Langroid
Langroid is an open-source Python framework designed to simplify LLM application development using a multi-agent programming paradigm. It treats agents as first-class citizens, allowing developers to configure agents with capabilities like LLMs, vector databases, and function-calling/tools. Langroid enables collaborative problem-solving through message exchange between agents, offering a "conversational programming" approach that works with both local/open and remote/proprietary LLMs.
Key Features
- Agent-Centric Architecture
- Task Delegation
- LLM Integration
- Vector Store Support
- Function Calling/Tools
- Caching
- Observability And Logging.
Use Cases
- Complex LLM Application Development
- Multi-Agent Collaboration
- Retrieval-Augmented Generation
- Chatbot Creation
- AI-Assisted Task Automation.
Similar Agents in AI Agents Frameworks
upsonicAI
Task-oriented AI agent framework for digital workers and vertical AI agents
Vectara-agentic
Build AI assistants and Agents with Vectara
LM-Kit SDK
Build Smarter Apps with Language Models
Smolagents AI Agent
Create AI agents with minimal code using Hugging Face's lightweight framework
Modelscope-Agent
ModelScope-Agent: An agent framework connecting models in ModelScope with the world
Rig
Build powerful, modular LLM applications in Rust with unified interfaces and high performance.