← Back to community benchmarks
gemma-4-12B-oQ8
Performance
64k
tokens
698.4
PP tok/s
17.4
TG tok/s
93842
TTFT (ms)
16.6
Peak mem (GB)
Hardware
Chip
M5 Pro (20c)
Memory
64 GB
GPU Cores
20
Software
oMLX
v0.4.2.dev2
macOS
macOS 26.5
Context
65,536