🖥️ AI Rig: 8x RTX 3090 (192GB VRAM) 256GB RAM · Threadripper 3975WX Daily driver: M1 Pro 🤖 I build local-first AI agents + LLM infra EXL3 · MoE offloading in open-source project-maya 💼 Open to paid agent & infra work. DMs open. 🎮 Also here for fitness, gaming, anime (Naruto forever), Marvel, music, movies, 3D printing, and getting into VR/AR dev.

one prompt, one file, a full asteroids game. running locally on my 8x 3090 rig qwen 3.8 flash-next on exllamav3 (6.05bpw + MTP draft) 124 tok/s 0.44s to first token ~1,270W while generating (8 cards, each capped at 160W) @Alibaba_Qwen @turboderp_

glad i bought this rig one piece at a time. pricing the whole thing at today's prices would be a very different conversation

I haven’t finished the new cyberpunk season 2 yet but I’m on episode 8 So far I like the first season better

glm 5.3 and qwen 3.8 already run at good quants on the new rig. deepseek is the boss fight

Trying out that GPU unlock

finished the 8x3090 build and now the thing holding it back is my house wiring. every card stays power limited until an electrician comes out

Not sure what I’ll get next a A6000 48GB or a DGX Spark or Ryzen Max

@loktar00 @axitek

Finally got my extra PSU cables to install the 8th 3090

favorite part of letting an agent run is reading the summary and finding out what it decided i wanted

night city in the rain is still the best gpu demo nobody had to make