▲ local intelligence

can it run

Can the Apple M4 Max run gemma3 12B?

FULL

9.6 GB needed of 34 GB usable — headroom for context

~34GB usable · Apple M4 Max · model file 8.1GB · 12B

~37–54 tok/s

computed roofline: 546 GB/s × 0.55–0.80 efficiency window / 8.1GB (Q4_K_M) — bands, never points.

Run gemma3 12B on instead

→ Tesla M60 (16GB, used) — cheapest catalog machine that runs gemma3 12B fully ($35)

→ every model the Apple M4 Max can run — full list

→explore the full catalog91 machines indexed · live prices · what each one can run
2026-10-01 · ← all models on the Apple M4 Max · fit = weights + 4k KV vs usable memory