can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB) run Laguna S 2.1 (MoE)?

Laguna · openmdw-1.1 · estimated for a 144.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Laguna S 2.1 (MoE) at FP16 · 235.2 GB · 32–48 tok/s on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB).

Every quantization on this machine

QuantSizeSpeedFit
Q3_K_M54.0 GB252–378 tok/sRuns great
Q4_K_M73.1 GB231–347 tok/sRuns great
Q5_K_M87.9 GB217–326 tok/sRuns great
Q6_K97.9 GB209–313 tok/sRuns great
Q8_0125.0 GB189–283 tok/sRuns great
FP16235.2 GB32–48 tok/sRuns great

Ran Laguna S 2.1 (MoE) on a NVIDIA GH200 Grace Hopper (144GB HBM + 480GB)? Add your real numbers →