Agent frameworks

Code libraries for AI agents that plan steps, call tools, remember context and hand work to other agents.

Search AirStack

Why teams use it

Who uses it: AI engineer, Software engineer, ML engineer, Research engineer, Back-end engineer. Watch for: Frameworks change quickly and add layers to debug; prefer the smallest one that does the job, and pin versions. AirStack’s guide to the category.

12 products

ProductWhat it says it doesStarting at
MastraMastra is an open-source TypeScript framework for building AI-powered applications and agents.Not observed
AgnoAgno is the self-driving platform for agents that do real work.Not observed
CrewAIThe Enterprise Agent Build & Runtime for the work your business runs onNot observed
Google ADKADK is the open-source agent development framework that lets you build, debug, and deploy reliable AI agents at enterprise scale.Not observed
Pydantic AIAgents, realtime voice, image generation, embeddings. Every model, every interface, typed end to end.Not observed
LangGraphAgent Orchestration Framework for Reliable AI AgentsNot observed
Semantic KernelIntegrate cutting-edge LLM technology quickly and easily into your appsNot observed
smolagentssmolagents is an open-source Python library designed to make it extremely easy to build and run agents using just a few lines of code.Not observed
Claude Agent SDKPython SDK for Claude Agent.Not observed
DSPyDSPy is a Python framework for building AI systems.Not observed
OpenAI Agents SDKThe OpenAI Agents SDK enables you to build agentic AI apps in a lightweight, easy-to-use package with very few abstractions.Not observed
AutoGenA framework for building AI agents and applicationsNot observed

Prices are list prices as written on each pricing page; hover one for its quote. “Not observed” means AirStack has not found it on that product’s pages.