local intelligence

can it run

Can the Apple M4 Pro run qwen2.5vl 32B?

TIGHT

23.0 of 22 GB usable — barely over; real with q8 KV cache (halves KV) or shorter context

~22GB usable · Apple M4 Pro · model file 22GB · 32B

~6.8–9.9 tok/s

computed roofline: 273 GB/s × 0.55–0.80 efficiency window / 22.0GB (Q4_K_M) — bands, never points.

Run qwen2.5vl 32B on instead

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

every model the Apple M4 Pro can run — full list

explore the full catalog84 machines indexed · live prices · what each one can run
2026-09-17 · ← all models on the Apple M4 Pro · fit = weights + 4k KV vs usable memory