← Back to community benchmarks

gemma-4-e4b-it

M4 Max (40c) · 128 GB · 8bit · 2026-05-14
Performance
64k
tokens
1,905
PP tok/s
44.9
TG tok/s
34411
TTFT (ms)
11.1
Peak mem (GB)
Hardware
Chip M4 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,396 71.8 9.0 GB view
4k 2,462 69.6 9.1 GB view
8k 2,435 67.0 9.2 GB view
16k 2,348 62.5 9.5 GB view
32k 2,180 55.4 10.2 GB view
64k 1,905 44.9 11.1 GB current
Batching Results
Batch Size TG tok/s Speedup
71.8 1.00×
137.7 1.92×
203.4 2.83×
233.1 3.25×