← Back to community benchmarks

gemma-4-E4B-it-PARO

M5 Max (40c) · 128 GB · 4bit · 2026-05-29
Performance
128k
tokens
3,197
PP tok/s
31.7
TG tok/s
40994
TTFT (ms)
14.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.12
macOS macOS 26.6
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,741 104.3 10.6 GB view
4k 4,681 101.4 10.6 GB view
64k 4,536 50.1 12.6 GB view
128k 3,197 31.7 14.9 GB current
Batching Results
Batch Size TG tok/s Speedup
104.3 1.00×
157.0 1.51×
227.9 2.19×