← Back to community benchmarks

gemma-4-E4B-it

M5 (10c) · 32 GB · 6bit · 2026-05-16
Performance
16k
tokens
1,695
PP tok/s
22.5
TG tok/s
9669
TTFT (ms)
8.4
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,659 25.4 7.9 GB view
4k 1,795 24.7 8.0 GB view
8k 1,776 24.0 8.2 GB view
16k 1,695 22.5 8.4 GB current
32k 1,531 19.9 8.9 GB view
64k 1,257 16.2 10.0 GB view
Batching Results
Batch Size TG tok/s Speedup
25.4 1.00×
50.4 1.98×
70.3 2.77×
76.7 3.02×