← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Pro (20c) · 64 GB · 4bit · 2026-04-28
Performance
8k
tokens
2,025
PP tok/s
60.9
TG tok/s
4045
TTFT (ms)
14.8
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.7
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,451 72.9 14.3 GB view
4k 1,956 70.0 14.7 GB view
8k 2,025 60.9 14.8 GB current
16k 1,982 53.2 15.2 GB view
32k 1,829 31.6 16.0 GB view
64k 1,438 14.5 17.7 GB view
Batching Results
Batch Size TG tok/s Speedup
72.9 1.00×
107.7 1.48×
134.1 1.84×
163.8 2.25×