← Back to community benchmarks

gemma-4-E4B-it-ultra-uncensored-heretic8bit-int8-affine

M3 Max (40c) · 64 GB · 8bit · 2026-05-11
Performance
16k
tokens
1,193
PP tok/s
24.8
TG tok/s
13738
TTFT (ms)
9.7
Peak mem (GB)
Hardware
Chip M3 Max (40c)
Memory 64 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,134 36.2 9.2 GB view
4k 1,282 32.8 9.3 GB view
8k 1,292 29.6 9.4 GB view
16k 1,193 24.8 9.7 GB current
32k 991.0 18.6 10.2 GB view
64k 902.5 21.3 10.8 GB view
128k 892.0 20.1 12.5 GB view
195k 780.0 14.4 14.3 GB view