← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-13
Performance
64k
tokens
2,672
PP tok/s
64.2
TG tok/s
24530
TTFT (ms)
18.5
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 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,218 113.7 14.3 GB view
8k 3,588 105.2 15.0 GB view
32k 3,357 83.0 16.3 GB view
64k 2,672 64.2 18.5 GB current
Batching Results
Batch Size TG tok/s Speedup
113.7 1.00×
153.2 1.35×
233.9 2.06×