ai agent frameworks

code-first agent frameworks ranked on production reliability, not github stars — with the durable-execution and low-code layers they sit next to made explicit.

5 listed · 5 researched by us · featured first, then most upvoted

LangGraph

we checked thisfree — MIT (open source)

the most production-proven stateful-agent runtime

#teams-that-need-durable-#human-in-the-loop#fine-grained-control

OpenAI Agents SDK

we checked thisfree — MIT (open source)

minimal, clean and production-ready — despite the name

#teams-that-want-a-small#readable-agent-sdk-with-

Pydantic AI

we checked thisfree — MIT (open source)

type-safe agents from the team that validates half of python ai

#python-teams-that-want-t#validated-agent-outputs#clean-observability

Google ADK

we checked thisfree — Apache-2.0 (open source)

code-first, ga, enterprise deploy — with gemini gravity

#teams-building-multi-age

Claude Agent SDK

we checked thisfree — MIT (open source)

the exact harness that runs claude code, as a library

#autonomous-agents-that-u#commands#code-with-long-running-c

this page is deliberately short: a handful of ai agent frameworks we looked at properly, rather than everything that exists. if you build one that belongs here, a free listing takes two minutes and $50 pins it to the top of this page.

add a tool →