← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 8bit · 2026-04-12
Performance
32k
tokens
2,520
PP tok/s
40.7
TG tok/s
13002
TTFT (ms)
29.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.5.dev2
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,141 84.5 28.2 GB view
4k 3,063 79.7 28.6 GB view
8k 3,356 72.9 28.7 GB view
16k 2,873 62.4 29.1 GB view
32k 2,520 40.7 29.9 GB current
64k 1,930 26.7 31.6 GB view
Batching Results
Batch Size TG tok/s Speedup
84.5 1.00×
123.3 1.46×
169.5 2.01×