Frameworks

Structured, comparable entries for agent orchestration frameworks and SDKs. Choose by fit — not by hype.

Framework	Category	Languages	Best for
AutoGen	multi-agent	Python	Conversational multi-agent experiments; Coding agents with reviewer/user-proxy patterns
CrewAI	multi-agent	Python	Role-oriented multi-agent prototypes; Ops and content workflows with specialist personas
DSPy	framework	Python	Teams that want optimizable LM pipelines; Research-to-production prompt programming
LangChain	framework	Python, JavaScript	Composable LLM app building blocks; RAG pipelines and tool integrations
LangGraph	orchestration	Python, JavaScript	Production agent control flows with explicit state; Human-in-the-loop nodes and durable runs
LlamaIndex	framework	Python, TypeScript	RAG-heavy and data-centric agents; Enterprise document workflows
Mastra	framework	TypeScript	TypeScript/full-stack JavaScript teams; Agents embedded in modern web backends
OpenAI Agents SDK	sdk	Python, TypeScript	OpenAI-model-centric agent apps; Handoffs between specialized agents
PydanticAI	framework	Python	Typed Python codebases that want structured agent I/O; Teams already invested in Pydantic
Semantic Kernel	sdk	C#, Python, Java	Enterprise Microsoft-oriented stacks; .NET applications adding agentic capabilities

AutoGen is a multi-agent framework for conversational collaboration between specialized agents, often used for coding and research workflows.

CrewAI is a Python framework for role-based multi-agent teams that collaborate on tasks with tools and processes.

DSPy is a framework for programming—not just prompting—LMs, with optimizers that tune pipelines and agent-like multi-step modules.

LangChain is a broad framework for building LLM applications with chains, tools, retrieval, and integrations used alongside agent orchestrators.

LangGraph is a library for building stateful, graph-based agent workflows with explicit control flow and durable execution patterns.

LlamaIndex is a framework specializing in data connectors, indexing, retrieval, and agentic workflows over private and enterprise data.

Mastra is a TypeScript framework for building AI applications and agents with workflows, tools, and modern JS/TS developer experience.

The OpenAI Agents SDK is a toolkit for building agent applications with tool use, handoffs, and guardrail-oriented patterns in the OpenAI ecosystem.

PydanticAI is a Python agent framework emphasizing type-safe structured outputs, dependency injection, and ergonomic agent definitions.

Semantic Kernel is Microsoft's SDK for orchestrating AI plugins, planners, and agents across enterprise .NET, Python, and Java stacks.