← Back to community benchmarks

gemma-4-31b-it

M4 Max (32c) · 36 GB · 4bit · 2026-05-22
Performance
32k
tokens
132.4
PP tok/s
13.7
TG tok/s
247552
TTFT (ms)
22.8
Peak mem (GB)
Hardware
Chip M4 Max (32c)
Memory 36 GB
GPU Cores 32
Software
oMLX v0.3.9
macOS macOS 26.5
Context 32,768