← Back to community benchmarks

gemma-4-31B-it-oQ6

M3 Ultra (60c) · 256 GB · 6bit · 2026-05-07
Performance
32k
tokens
222.7
PP tok/s
17.7
TG tok/s
147113
TTFT (ms)
32.5
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.8
macOS macOS 26.3
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 244.7 22.2 25.3 GB view
4k 248.2 21.4 27.9 GB view
8k 246.5 20.8 28.4 GB view
16k 238.6 17.7 29.6 GB view
32k 222.7 17.7 32.5 GB current
64k 194.5 13.1 38.5 GB view
Batching Results
Batch Size TG tok/s Speedup
22.2 1.00×
30.5 1.37×
34.5 1.55×
38.5 1.73×