There’s still some final touches but this thing has been reliably crunching numbers for 3 days now. This thing is solid 🔥🔥
Running Prem Ai with Vicuna 3b, Whisper, and Stable Diffusion. All are cruising nicely. So cool having my own Ai to work with. Exploring some coding specific models tomorrow. https://v.nostr.build/azoQ.mov
Looks nice!
What's the specs on that beast?
Since I have some serious space constraints I went for a used laptop with i7 CPU/64GB, with a mobile RTX 5000 / 16GB VRAM. Not the fastest, but combined with a used 3090 24GB eGPU it can load quite big models.
All for about 2600 USD.
