← Back to community benchmarks

gemma-4-31B-it

M5 Max (40c) · 128 GB · 8bit · 2026-05-16
Performance
32k
tokens
333.0
PP tok/s
4.7
TG tok/s
98417
TTFT (ms)
39.3
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 615.8 15.3 31.8 GB view
4k 531.3 14.6 34.5 GB view
8k 505.6 13.2 35.0 GB view
16k 446.9 11.5 36.1 GB view
32k 333.0 4.7 39.3 GB current
64k 210.7 5.7 45.6 GB view
Batching Results
Batch Size TG tok/s Speedup
15.3 1.00×
13.1 0.86×
18.4 1.20×
22.8 1.49×