← Back to community benchmarks

gemma-4-e4b

M4 (10c) · 16 GB · 8bit · 2026-04-15
Performance
16k
tokens
604.7
PP tok/s
17.1
TG tok/s
27097
TTFT (ms)
8.3
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 16 GB
GPU Cores 10
Software
oMLX v0.3.5.dev1
macOS macOS 26.3
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 531.4 19.6 7.9 GB view
4k 622.0 19.3 8.0 GB view
8k 624.7 18.8 8.1 GB view
16k 604.7 17.1 8.3 GB current
Batching Results
Batch Size TG tok/s Speedup
19.6 1.00×
38.1 1.94×
61.3 3.13×