← Back to community benchmarks

gemma-4-26B-A4B-it

M5 Max (40c) · 128 GB · 8bit · 2026-06-01
Performance
64k
tokens
1,695
PP tok/s
40.2
TG tok/s
38658
TTFT (ms)
30.0
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.12
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,747 68.8 25.8 GB view
4k 2,659 63.2 26.4 GB view
8k 2,944 67.2 26.6 GB view
16k 2,683 61.3 27.1 GB view
32k 2,251 53.1 28.0 GB view
64k 1,695 40.2 30.0 GB current
Batching Results
Batch Size TG tok/s Speedup
68.8 1.00×
102.9 1.50×
121.3 1.76×