← Back to community benchmarks

gemma-4-12B-it-OptiQ

M3 Ultra (60c) · 256 GB · 4bit · 2026-07-21
Performance
64k
tokens
492.3
PP tok/s
42.5
TG tok/s
133114
TTFT (ms)
13.8
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.5.1
macOS macOS 26.5.2
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 708.2 23.6 9.3 GB view
4k 664.3 52.0 9.7 GB view
8k 633.0 51.6 10.6 GB view
16k 597.5 49.8 10.7 GB view
32k 558.4 48.5 11.6 GB view
64k 492.3 42.5 13.8 GB current
Batching Results
Batch Size TG tok/s Speedup
23.6 1.00×
78.2 3.31×
138.1 5.85×
159.7 6.77×