← Back to community benchmarks

gemma-4-E4B-it-PARO

M5 Max (40c) · 128 GB · 4bit · 2026-05-25
Performance
1k
tokens
2,871
PP tok/s
111.6
TG tok/s
357
TTFT (ms)
10.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.10
macOS macOS 26.5
Context 1,024
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,871 111.6 10.6 GB current
64k 4,791 58.2 12.6 GB view
128k 3,575 40.2 14.9 GB view
Batching Results
Batch Size TG tok/s Speedup
111.6 1.00×
194.0 1.74×
305.6 2.74×