← Back to community benchmarks

gemma-4-e4b-it

M1 Max (24c) · 32 GB · 8bit · 2026-04-17
Performance
32k
tokens
635.5
PP tok/s
23.0
TG tok/s
51565
TTFT (ms)
9.0
Peak mem (GB)
Hardware
Chip M1 Max (24c)
Memory 32 GB
GPU Cores 24
Software
oMLX v0.3.5
macOS macOS 26.3
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 550.4 45.2 8.0 GB view
4k 655.5 42.3 8.1 GB view
8k 667.3 39.6 8.2 GB view
16k 663.6 31.4 8.4 GB view
32k 635.5 23.0 9.0 GB current
Batching Results
Batch Size TG tok/s Speedup
45.2 1.00×
71.0 1.57×
86.1 1.90×