← Back to community benchmarks

gemma-4-E4B-it-heretic-oQ6

M3 (10c) · 16 GB · 6bit · 2026-05-18
Performance
8k
tokens
496.6
PP tok/s
42.4
TG tok/s
16496
TTFT (ms)
7.2
Peak mem (GB)
Hardware
Chip M3 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.3.9.dev2
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 523.5 48.8 6.9 GB view
4k 527.5 45.2 7.0 GB view
8k 496.6 42.4 7.2 GB current
16k 433.1 29.7 7.4 GB view
32k 237.7 17.2 8.0 GB view
64k 190.9 10.7 9.2 GB view
Batching Results
Batch Size TG tok/s Speedup
48.8 1.00×
14.8 0.30×
19.7 0.40×
22.2 0.45×