← Back to community benchmarks

gemma-4-31b-it

M5 Max (40c) · 128 GB · 6bit · 2026-04-03
Performance
16k
tokens
551.1
PP tok/s
11.5
TG tok/s
29730
TTFT (ms)
28.7
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.2
macOS macOS 26.4
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 643.2 16.9 26.0 GB view
4k 630.4 16.1 27.7 GB view
8k 568.4 14.6 27.9 GB view
16k 551.1 11.5 28.7 GB current
32k 509.8 8.2 31.0 GB view
64k 405.3 5.4 35.7 GB view
128k 268.8 2.9 45.3 GB view
Batching Results
Batch Size TG tok/s Speedup
16.9 1.00×
19.3 1.14×
21.1 1.25×