I want run open source LLMs on this prebuilt AI. Anyone tested this before? (autonomous.ai)

seems a tad expensive for (nerfed) consumer-grade GPUs IMO

cool case, i'll go for opt 2 as they have Threadripper.

good point is that this integrates some existing models for you to use. You can test different models with this. But still need to upgrade if you wanna try larger models. The more Vram the better

It reminds me of a similar product [0] from George Hotz

[0] https://tinygrad.org/#tinybox