← Back to community benchmarks

gemma-4-31B

M5 Max (40c) · 128 GB · unknown · 2026-05-22
Performance
16k
tokens
675.9
PP tok/s
7.2
TG tok/s
24241
TTFT (ms)
63.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 562.7 7.4 58.9 GB view
4k 753.3 7.3 61.5 GB view
16k 675.9 7.2 63.5 GB current
64k 459.9 5.8 70.7 GB view