local intelligence

can it run

Can the NVIDIA L40S 48GB run DEEPSEEK-R1 distill 70B?

TIGHT

44.2 of 48 GB usable — keep context modest

48GB · Ada Lovelace · model file 43GB · distill 70B

~9.2–15 tok/s

computed roofline: 864 GB/s × 0.46–0.76 efficiency window / 43.0GB (Q4_K_M) — bands, never points.

Run DEEPSEEK-R1 distill 70B on instead

NIMO AI Mini PC (128GB) — cheapest catalog machine that runs DEEPSEEK-R1 distill 70B fully ($1999)

rtx-pro-6000 — the next step up from NVIDIA L40S 48GB

every model the NVIDIA L40S 48GB 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 L40S 48GB · fit = weights + 4k KV vs usable memory