The most widely used framework for building LLM applications.
Frameworks for Multi-agent
The 53 frameworks in the runchr registry that do multi-agent work. Coordinate several agents on one job.
50 of 53 tools · Metrics as of 2026-11-06
Agent meta-harness for Claude Code and Codex with swarms and memory
Organize role-playing agents like a small software company.
Microsoft's framework for multi-agent conversation.
Assemble a team of role-playing agents in a few lines.
Open-source book on AI agent design with 109 hands-on experiment projects.
AI-oriented quant investment platform for ML-driven research and production.
Lightweight self-hosted Python framework for building personal AI agents.
Model an agent workflow as a graph with state and checkpoints.
Frontend stack for building agent-powered apps with chat and generative UI.
Open-source drivers, sandboxes, and benchmarks for computer-use agents.
OpenAI's lightweight agent framework with handoffs built in.
A TypeScript framework for agents that fits a normal web stack.
Python framework for building RAG pipelines and agent workflows.
Rust-based agent operating system for running autonomous scheduled agents.
Open-source Python framework for building and researching multi-agent systems.
Open-source curriculum teaching MCP fundamentals with cross-language examples.
Open-source platform and SDK for building and deploying LLM chatbots and agents.
Build production AI agents and multi-agent workflows in Python and .NET.
100-line minimalist LLM framework for building agents and workflows.
Meta-harness orchestrating Claude Code, Codex, Cursor and custom agents.
Open-source runtime for colonies of parallel AI agents that do real work.
TypeScript framework for building agents with memory, tools, workflows, and MCP.
TypeScript framework for building MCP servers, apps, and agent clients.
Wrap any agent handler to give it identity, A2A messaging, and x402 payments.
Open-source Python framework for building private, offline-capable AI agents.
Open-source Python and TypeScript SDK for building production AI agents.
Self-evolving engine that turns agent experience into auditable Genes
TypeScript harness framework for building autonomous agents with sandboxes
Python framework for building autonomous and traditional AI agents.
TypeScript framework for building agentic apps with shared actions and UI.
Self-hosted TypeScript runtime for coordinating multi-agent teams.
Governance toolkit for AI agents: policy, identity, sandboxing, and reliability.
Self-hosted cloud agent cluster for running a collaborative team of AI agents.
Java framework for building distributed, production-grade, long-running agents.
Framework turning coding models into browser and desktop agents via a terminal.
Python SDK for monitoring AI agents, tracking LLM costs, and benchmarking.
Framework for training language agents with reinforcement learning.
Java framework for building multi-agent AI apps with RAG and workflows.
Open-source agent runtime with ReAct and Agent Team modes plus a terminal TUI.
Open-source framework for building and orchestrating multi-agent AI systems.
Plan-first agent framework for OpenCode that enforces your coding patterns.
JVM agent framework for building LLM agents in Kotlin and Java with Spring.
Durable Work layer keeping one task moving across Codex, Claude, and other
Ruby framework for building AI agents with tools, MCP, and 23 providers.
Python framework for building multi-agent LLM applications with tools and RAG.
AI agent framework for black-box pentesting, bug bounty, and red-team workflows.
Framework for running a 24/7 autonomous AI company with 14 agent roles.
Open source .NET framework for building multi-agent AI applications.
Python framework for building reliable GenAI agent backends with structured