← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 6bit · 2026-05-05
Performance
64k
tokens
2,534
PP tok/s
53.5
TG tok/s
25859
TTFT (ms)
23.4
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,748 96.4 20.0 GB view
4k 3,033 93.1 20.7 GB view
64k 2,534 53.5 23.4 GB current
128k 1,630 37.7 27.0 GB view
Batching Results
Batch Size TG tok/s Speedup
96.4 1.00×
133.8 1.39×
179.8 1.87×
220.4 2.29×