← Back to community benchmarks

gemma-4-e4b-it

M4 Pro (16c) · 48 GB · 8bit · 2026-05-25
Performance
8k
tokens
1,010
PP tok/s
39.3
TG tok/s
8115
TTFT (ms)
9.2
Peak mem (GB)
Hardware
Chip M4 Pro (16c)
Memory 48 GB
GPU Cores 16
Software
oMLX v0.3.10
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,003 41.7 9.0 GB view
4k 1,007 40.6 9.1 GB view
8k 1,010 39.3 9.2 GB current
16k 976.0 37.1 9.5 GB view
32k 904.3 33.4 10.2 GB view
64k 781.3 27.2 11.1 GB view
128k 617.8 20.8 13.3 GB view
195k 502.5 16.4 16.2 GB view
Batching Results
Batch Size TG tok/s Speedup
41.7 1.00×
78.3 1.88×
95.9 2.30×
102.1 2.45×