← Back to community benchmarks

gemma-4-E4B-it

M5 Max (40c) · 128 GB · 4bit · 2026-07-17
Performance
16k
tokens
2,709
PP tok/s
37.6
TG tok/s
6048
TTFT (ms)
6.6
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.5.1
macOS macOS 27.0
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 4,353 40.7 6.1 GB view
4k 2,401 44.8 6.2 GB view
8k 2,768 40.6 6.4 GB view
16k 2,709 37.6 6.6 GB current
32k 2,471 33.2 7.3 GB view
64k 2,087 26.3 8.2 GB view
Batching Results
Batch Size TG tok/s Speedup
40.7 1.00×
88.7 2.18×
138.6 3.41×