local intelligence

can it run

Can the NVIDIA RTX 3090 run qwen2.5vl 32B?

TIGHT

23.0 of 24 GB usable — keep context modest

24GB · Ampere GA102 · model file 22GB · 32B

~20–32 tok/s

computed roofline: 936 GB/s × 0.46–0.76 efficiency window / 22.0GB (Q4_K_M) — bands, never points.

Community check: ~26 tok/s · 30-33B Q4 · 20 runs · inside our computed band — localmaxxing.com

Run qwen2.5vl 32B on instead

Mac mini M5 Pro (64GB) — cheapest catalog machine that runs qwen2.5vl 32B fully ($1669)

rtx-5090 — the next step up from NVIDIA RTX 3090

every model the NVIDIA RTX 3090 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 RTX 3090 · fit = weights + 4k KV vs usable memory