← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 8bit · 2026-05-10
Performance
16k
tokens
3,343
PP tok/s
74.5
TG tok/s
4902
TTFT (ms)
27.1
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 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,047 88.1 25.8 GB view
4k 3,103 84.3 26.4 GB view
8k 3,435 80.9 26.6 GB view
16k 3,343 74.5 27.1 GB current
32k 2,987 63.4 27.5 GB view
64k 2,351 49.5 29.2 GB view
Batching Results
Batch Size TG tok/s Speedup
88.1 1.00×
125.2 1.42×
176.8 2.01×
226.4 2.57×