← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-21
Performance
16k
tokens
3,322
PP tok/s
95.8
TG tok/s
4932
TTFT (ms)
15.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.9rc1
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,144 116.1 14.3 GB view
4k 3,273 110.5 14.9 GB view
8k 3,647 104.2 15.1 GB view
16k 3,322 95.8 15.6 GB current
32k 2,813 81.6 16.5 GB view
64k 2,136 59.0 18.5 GB view
128k 1,324 33.1 22.7 GB view
195k 594.3 20.9 27.7 GB view
Batching Results
Batch Size TG tok/s Speedup
116.1 1.00×
124.8 1.07×
127.1 1.09×
153.5 1.32×