← Back to community benchmarks

gemma-4-E4B-it

M3 Ultra (60c) · 256 GB · unknown · 2026-06-02
Performance
32k
tokens
2,340
PP tok/s
40.2
TG tok/s
14003
TTFT (ms)
15.8
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.12
macOS macOS 26.5
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
16k 2,519 47.2 15.1 GB view
32k 2,340 40.2 15.8 GB current
64k 2,058 31.4 16.9 GB view