← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 8bit · 2026-05-25
Performance
16k
tokens
2,195
PP tok/s
70.7
TG tok/s
7463
TTFT (ms)
27.2
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.10
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,133 82.7 25.8 GB view
4k 1,782 82.3 26.5 GB view
8k 2,071 77.9 26.8 GB view
16k 2,195 70.7 27.2 GB current
32k 2,147 61.0 28.2 GB view
64k 1,705 42.4 30.2 GB view
Batching Results
Batch Size TG tok/s Speedup
82.7 1.00×
128.8 1.56×
135.2 1.63×