← Back to community benchmarks

gemma-4-E4B-it

M3 Max (30c) · 36 GB · 8bit · 2026-05-27
Performance
128k
tokens
896.1
PP tok/s
19.1
TG tok/s
146274
TTFT (ms)
13.3
Peak mem (GB)
Hardware
Chip M3 Max (30c)
Memory 36 GB
GPU Cores 30
Software
oMLX v0.3.11
macOS macOS 26.5
Context 131,072
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,532 43.3 8.9 GB view
4k 1,567 40.3 9.1 GB view
64k 1,194 28.0 10.5 GB view
128k 896.1 19.1 13.3 GB current
Batching Results
Batch Size TG tok/s Speedup
43.3 1.00×
62.8 1.45×
106.7 2.46×