local intelligence

can it run

Can the NVIDIA B200 192GB run qwen3.8 27B?

FULL

18.5 GB needed of 192 GB usable — headroom for context

192GB · Blackwell · model file 16.5GB · 27B

~220–370 tok/s

computed roofline: 8000 GB/s × 0.46–0.76 efficiency window / 16.5GB (Q4_K_M) — bands, never points.

Run qwen3.8 27B on instead

Tesla V100 32GB (used) — cheapest catalog machine that runs qwen3.8 27B fully ($640)

NVIDIA DGX Station (GB300) — upgrade path machine

every model the NVIDIA B200 192GB can run — full list

explore the full catalog90 machines indexed · live prices · what each one can run
2026-09-24 · ← all models on the NVIDIA B200 192GB · fit = weights + 4k KV vs usable memory