← Back to community benchmarks

gemma-4-31b-it

M5 Max (40c) · 128 GB · 4bit · 2026-04-02
Performance
32k
tokens
404.0
PP tok/s
8.6
TG tok/s
81107
TTFT (ms)
23.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.1
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 373.2 18.8 18.9 GB view
4k 446.6 18.2 20.5 GB view
8k 488.9 15.9 20.7 GB view
16k 469.4 12.5 21.5 GB view
32k 404.0 8.6 23.9 GB current
64k 308.3 5.5 28.6 GB view
Batching Results
Batch Size TG tok/s Speedup
18.8 1.00×
20.5 1.09×
28.4 1.51×
32.6 1.73×