← Back to community benchmarks

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

M5 Max (40c) · 128 GB · 4bit · 2026-04-05
Performance
8k
tokens
3,638
PP tok/s
98.4
TG tok/s
2252
TTFT (ms)
14.5
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.4
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,265 119.7 14.0 GB view
4k 3,338 110.5 14.5 GB view
8k 3,638 98.4 14.5 GB current
Batching Results
Batch Size TG tok/s Speedup
119.7 1.00×
177.2 1.48×
248.1 2.07×