← Back to community benchmarks

gemma-4-26B-A4B-it-uncensored-oQ8

M4 Pro (20c) · 64 GB · 8bit · 2026-05-01
Performance
128k
tokens
360.8
PP tok/s
1.1
TG tok/s
363266
TTFT (ms)
32.9
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.3
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 674.7 45.0 25.8 GB view
4k 723.7 44.1 26.4 GB view
8k 706.4 42.2 26.6 GB view
16k 619.0 37.1 27.1 GB view
32k 561.4 32.4 27.5 GB view
64k 480.5 12.2 29.2 GB view
128k 360.8 1.1 32.9 GB current
Batching Results
Batch Size TG tok/s Speedup
45.0 1.00×
65.0 1.44×