Could someone help me with the most popular agentic LLM frameworks such as AutoGPT, AutoGen, Crew AI, LangGraph, etc.?
I'm interested in setting them up locally with the most efficient configuration available. Have there been recent updates that enable them to run on top models like Phi-3 or LLAMA3 via local AI, etc.?
I'm particularly keen on options and comparisons, considering that I operate on a machine with three RTX 6000 GPUs (Ada).
What would be the best option/setup for optimizing performance with this setup?