← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Pro (20c) · 48 GB · 8bit · 2026-07-13
Performance
16k
tokens
1,640
PP tok/s
39.5
TG tok/s
9993
TTFT (ms)
27.5
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 48 GB
GPU Cores 20
Software
oMLX v0.5.0rc1
macOS macOS 26.5.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,779 46.6 25.8 GB view
4k 1,835 43.9 26.2 GB view
8k 1,738 42.9 27.1 GB view
16k 1,640 39.5 27.5 GB current
32k 1,449 34.5 28.9 GB view
64k 1,246 29.8 30.2 GB view
Batching Results
Batch Size TG tok/s Speedup
46.6 1.00×
81.1 1.74×