← Back to community benchmarks

gemma-4-e4b-it

M4 (10c) · 24 GB · 4bit · 2026-05-30
Performance
16k
tokens
506.1
PP tok/s
28.6
TG tok/s
32376
TTFT (ms)
5.9
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 24 GB
GPU Cores 10
Software
oMLX v0.3.12
macOS macOS 26.2
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 561.7 32.0 5.4 GB view
4k 557.7 31.4 5.5 GB view
8k 536.6 30.3 5.7 GB view
16k 506.1 28.6 5.9 GB current
32k 463.6 24.5 6.5 GB view
64k 388.1 19.2 7.6 GB view
Batching Results
Batch Size TG tok/s Speedup
32.0 1.00×
53.0 1.66×
62.4 1.95×
68.2 2.13×