Ask HN: HotPin – lossless 120B MoE inference on 24GB RAM (CPU, 50 loc)

By LozzKappa · 2026-07-25 · 2 points · 0 comments

I'm a mechatronics designer with a background in control systems, robotics, PCB design, and embedded hardware. I design physical systems: motors, sensors, microcontrollers, and real-time control loops. I applied this design thinking to LLM memory management – and it worked. HotP…

Open the full discussion on BetterNews