← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-28
Performance
64k
tokens
2,580
PP tok/s
59.3
TG tok/s
25401
TTFT (ms)
17.7
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
64k 2,580 59.3 17.7 GB current
128k 1,738 40.1 21.2 GB view
195k 1,293 29.1 24.9 GB view