← Back to community benchmarks

gemma-4-31b-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-19
Performance
195k
tokens
270.5
PP tok/s
7.9
TG tok/s
739387
TTFT (ms)
56.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9rc1
macOS macOS 26.4.1
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 687.8 23.9 16.7 GB view
4k 733.4 26.3 19.3 GB view
8k 682.3 25.0 19.8 GB view
16k 602.1 22.5 21.0 GB view
32k 546.9 19.2 23.9 GB view
64k 467.9 15.2 29.9 GB view
128k 346.9 10.4 41.9 GB view
195k 270.5 7.9 56.9 GB current
Batching Results
Batch Size TG tok/s Speedup
23.9 1.00×
28.2 1.18×
45.2 1.89×
52.3 2.19×