CrewAI
Role-playing agents working as a crew.
Stateful, cyclic agent graphs for production.
Local or cloud · No GPU
Pure runtime orchestration; no GPU. Backend can be any LLM.
Source is public — you can audit it, fork it, and you'll never lose access to your workflows if LangGraph the company changes direction.
Doesn't require a dedicated GPU — useful on laptops and headless servers.
Native Metal / MPS support — runs on M-series Macs without CUDA gymnastics.
Power-user score 90/100 — consistently rated highly by people who use this every day, not just benchmark chasers.
This page summarizes upstream documentation, release information, and editorially reviewed catalogue fields. It is not presented as a hands-on benchmark. Verify changing requirements at the official project; report stale data through our corrections channel.
Runs CPU-only — no CUDA / driver gymnastics required.
No dedicated GPU required.
Open-source AND ships an API — easy to integrate, possible to host yourself.
Catalogue entry last updated 57 days ago — re-verification due soon.
Three picks across different tradeoffs — so you don't end up with three near-clones of LangGraph.
LangGraph is LangChain's answer to "agents are actually state machines." You define nodes (LLM calls, tool calls) and edges (transitions), and the framework gives you persistence, time-travel debugging, and human-in-the-loop interrupts. The serious choice if you're putting agents in production.
OSS Python/JS library free. LangGraph Platform offers paid hosted deployment.