← Back to community benchmarks

gemma-4-26B-A4B-it-OptiQ

M3 Ultra (60c) · 256 GB · 4bit · 2026-05-16
Performance
16k
tokens
1,658
PP tok/s
76.6
TG tok/s
9885
TTFT (ms)
18.5
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.8
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,248 92.4 17.2 GB view
4k 1,571 86.3 17.9 GB view
8k 1,701 82.6 18.0 GB view
16k 1,658 76.6 18.5 GB current
32k 1,507 63.7 19.4 GB view
64k 1,237 46.5 20.6 GB view
128k 888.4 30.1 24.2 GB view
Batching Results
Batch Size TG tok/s Speedup
92.4 1.00×
135.8 1.47×
184.9 2.00×