← Back to community benchmarks

gemma-4-12B-it-OptiQ

M3 Ultra (60c) · 256 GB · 4bit · 2026-07-21
Performance
4k
tokens
651.6
PP tok/s
51.5
TG tok/s
6286
TTFT (ms)
10.0
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 4,096
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 705.1 51.8 9.3 GB view
4k 651.6 51.5 10.0 GB current
8k 638.6 50.8 10.3 GB view
16k 598.8 50.8 11.1 GB view
32k 554.4 48.3 11.7 GB view
64k 491.4 44.3 13.5 GB view
Batching Results
Batch Size TG tok/s Speedup
51.8 1.00×
57.5 1.11×
147.6 2.85×
152.8 2.95×