local intelligence

can it run

Can the NVIDIA GeForce RTX 4070 Ti Super run devstral 24B?

TIGHT

15.2 of 16 GB usable — keep context modest

16GB · Ada · model file 14.6GB · 24B

~25–37 tok/s

computed roofline: 672 GB/s × 0.54–0.80 efficiency window / 14.6GB (Q4_K_M) — bands, never points.

Run devstral 24B on instead

Tesla P40 (24GB, used) — cheapest catalog machine that runs devstral 24B fully ($290)

rtx-4090 — the next step up from NVIDIA GeForce RTX 4070 Ti Super

every model the NVIDIA GeForce RTX 4070 Ti Super can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the NVIDIA GeForce RTX 4070 Ti Super · fit = weights + 4k KV vs usable memory