← Back to community benchmarks

gemma-4-e4b-it

M5 Pro (20c) · 64 GB · 4bit · 2026-05-19
Performance
64k
tokens
2,966
PP tok/s
43.5
TG tok/s
22099
TTFT (ms)
7.1
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.5
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 3,903 80.8 5.6 GB view
4k 3,761 77.0 5.7 GB view
8k 3,971 73.7 5.8 GB view
16k 3,928 67.2 6.0 GB view
32k 3,568 56.0 6.4 GB view
64k 2,966 43.5 7.1 GB current
128k 2,140 28.4 8.8 GB view
195k 1,572 20.9 10.7 GB view
Batching Results
Batch Size TG tok/s Speedup
80.8 1.00×
140.0 1.73×
177.6 2.20×
205.2 2.54×