← Back to community benchmarks

gemma-4-E4B-it-oQ6

M4 (10c) · 16 GB · 6bit · 2026-05-05
Performance
8k
tokens
620.0
PP tok/s
22.2
TG tok/s
13214
TTFT (ms)
7.0
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 608.4 22.9 6.7 GB view
4k 629.9 22.8 6.8 GB view
8k 620.0 22.2 7.0 GB current
16k 592.2 21.0 7.2 GB view
32k 518.9 17.9 7.5 GB view
Batching Results
Batch Size TG tok/s Speedup
22.9 1.00×
42.3 1.85×
48.2 2.10×
50.4 2.20×