$2,500 Local Rig Runs 27B Model at 150 tokens/s

Developer casperhansen shared a $2,500 local inference build (used RTX 3090, 64GB DDR5, Ryzen) that runs a 27B model at 150 tokens/s and squeezes the 328GB GLM-5.3-Flash down to 125GB for local execution.

2026-11-05 ~ 2026-11-05 · 2 related posts

• A $2.5k RTX 3090 box runs 328GB GLM-5.3-Flash locally via expert offloading — casper_hansen_ · 2026-11-05
• A $2.5k local rig runs Qwen3.8 27B at 150 tokens/s, benchmarks show — casper_hansen_ · 2026-11-05