← Back to community benchmarks

gemma-4-26B-A4B-it-oQ4

M4 (10c) · 32 GB · 4bit · 2026-04-17
Performance
16k
tokens
348.2
PP tok/s
15.6
TG tok/s
47058
TTFT (ms)
15.3
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.6
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 366.6 30.9 14.4 GB view
4k 377.5 29.6 14.8 GB view
8k 369.7 20.8 14.9 GB view
16k 348.2 15.6 15.3 GB current
32k 309.5 9.9 16.1 GB view
64k 249.8 6.4 17.8 GB view
Batching Results
Batch Size TG tok/s Speedup
30.9 1.00×
46.0 1.49×
55.1 1.78×
64.5 2.09×