local intelligence

can it run

Can the NVIDIA GeForce RTX 4090 run devstral 24B?

FULL

15.2 GB needed of 24 GB usable — headroom for context

24GB · Ada · model file 14.6GB · 24B

~32–52 tok/s

computed roofline: 1008 GB/s × 0.46–0.76 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-5090 — the next step up from NVIDIA GeForce RTX 4090

every model the NVIDIA GeForce RTX 4090 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 4090 · fit = weights + 4k KV vs usable memory