← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-08
Performance
64k
tokens
2,568
PP tok/s
59.5
TG tok/s
25521
TTFT (ms)
18.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,175 114.7 15.5 GB view
4k 3,413 111.0 16.2 GB view
8k 3,696 104.0 16.3 GB view
16k 3,695 94.9 16.8 GB view
32k 3,279 79.0 17.2 GB view
64k 2,568 59.5 18.9 GB current
128k 1,714 40.0 22.5 GB view
195k 1,189 26.8 26.2 GB view
Batching Results
Batch Size TG tok/s Speedup
114.7 1.00×
174.5 1.52×
223.6 1.95×
281.3 2.45×