← Back to community benchmarks

gemma-4-31B-it

M4 Max (40c) · 128 GB · unknown · 2026-07-13
Performance
16k
tokens
119.5
PP tok/s
4.4
TG tok/s
137048
TTFT (ms)
73.5
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.5.1
macOS macOS 26.5.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 219.3 7.4 58.5 GB view
4k 177.2 6.2 60.6 GB view
8k 144.7 4.2 64.1 GB view
16k 119.5 4.4 73.5 GB current
32k 122.3 4.5 85.6 GB view
64k 99.8 4.1 113.1 GB view
Batching Results
Batch Size TG tok/s Speedup
7.4 1.00×
4.8 0.65×
9.4 1.27×
17.5 2.36×