Skip to content

AI & Agent Frameworks

Software frameworks, SDKs, and libraries for developing AI-driven applications, multi-agent systems, structured LLM outputs, and model training pipelines.

Contents

Framework What it does
AG2 Next-generation AutoGen framework evolution for autonomous multi-agent systems
AutoGen Microsoft framework for building multi-agent conversational systems
AutoGen Studio No-code/low-code web UI for designing, testing, and debugging AutoGen multi-agent workflows
AWS Kiro AWS serverless framework for event-driven agent orchestration
Axolotl Streamlined LLM fine-tuning framework supporting QLoRA and FlashAttention-3
CrewAI Framework for orchestrating role-playing autonomous AI agents
Distilabel Synthetic data generation and AI feedback pipeline framework by Argilla
DSPy Declarative framework for compiling and optimizing LLM prompts and pipelines
Embabel Agent Framework Kotlin/Java Spring Boot framework for enterprise LLM agent integration
FastAPI High-performance Python web framework widely used for serving AI microservices
Firebase Genkit Google open-source framework for building AI features in Node.js and Go
Google ADK Agent Development Kit for building, testing, and deploying Gemini agentic workflows
GraphRAG Graph-based retrieval augmented generation framework for multi-hop reasoning
Haystack Open-source NLP and RAG framework by Deepset for building search and agent pipelines
Instructor Python/TypeScript library for structured LLM outputs powered by Pydantic
Langflow Visual framework for building multi-agent and RAG applications
LangGraph Stateful multi-agent orchestration framework built on LangChain primitives
LLaMA-Factory Unified framework for fine-tuning 100+ LLMs with intuitive web interface
MageVL Multimodal vision-language agent framework for fine-grained image understanding
Mastra TypeScript agent framework with built-in workflows, RAG, and local evaluation
Microsoft Agent Framework Enterprise multi-agent runtime and governance framework by Microsoft
Microsoft Agent Framework Harness Testing and execution harness for Microsoft Agent Framework applications
Mycelium Decentralized peer-to-peer AI agent interaction and routing framework
NeMo AutoModel NVIDIA framework for automated model selection, quantization, and deployment
OpenAI Agents SDK Official lightweight SDK for building multi-agent workflows with OpenAI models
PydanticAI Production-grade Python agent framework built by the Pydantic team
Rivet Visual node-based AI programming environment and remote debugging tool
Semantic Kernel Microsoft SDK for integrating LLMs into C#, Python, and Java applications
Smolagents Bare-bones, code-first Python agent library by Hugging Face
Superinterface Developer platform and UI framework for embedding AI assistants in web apps
Tritium High-speed LLM orchestration runtime optimized for low-latency agent loops

  • Last reviewed: 2027-01-07