← Back to community benchmarks

gemma-4-31b-it

M4 Max (40c) · 128 GB · 4bit · 2026-04-03
Performance
16k
tokens
187.5
PP tok/s
15.2
TG tok/s
87402
TTFT (ms)
22.3
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.1
macOS macOS 15.7.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 199.9 20.0 18.9 GB view
4k 197.8 18.9 20.5 GB view
8k 195.1 17.4 21.1 GB view
16k 187.5 15.2 22.3 GB current
32k 174.1 2.1 24.6 GB view
64k 152.5 0.4 29.4 GB view
128k 120.0 0.3 38.8 GB view
Batching Results
Batch Size TG tok/s Speedup
20.0 1.00×
30.8 1.54×