← Back to community benchmarks

gemma-4-E4B-it

M5 (10c) · 32 GB · 8bit · 2026-06-05
Performance
8k
tokens
1,925
PP tok/s
21.4
TG tok/s
4256
TTFT (ms)
9.0
Peak mem (GB)
Hardware
Chip M5 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.4.1
macOS macOS 26.5.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,788 22.6 8.8 GB view
4k 1,951 21.6 8.9 GB view
8k 1,925 21.4 9.0 GB current
16k 1,848 19.4 9.3 GB view
32k 1,635 17.8 9.8 GB view
64k 1,317 14.9 10.9 GB view
128k 957.2 11.5 13.2 GB view
Batching Results
Batch Size TG tok/s Speedup
22.6 1.00×
46.6 2.06×
77.9 3.45×