← Back to community benchmarks

gemma-4-31B-it

M5 Max (40c) · 128 GB · 6bit · 2026-05-13
Performance
32k
tokens
300.2
PP tok/s
9.5
TG tok/s
109141
TTFT (ms)
36.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 563.9 15.1 29.5 GB view
4k 527.8 14.4 32.2 GB view
8k 514.6 13.9 32.7 GB view
16k 447.9 11.5 33.9 GB view
32k 300.2 9.5 36.7 GB current
64k 258.4 7.8 42.8 GB view
Batching Results
Batch Size TG tok/s Speedup
15.1 1.00×
2.5 0.17×
13.8 0.91×