← Back to community benchmarks

gemma-4-26B-A4B-it-oQ3

M2 Pro (19c) · 32 GB · 3bit · 2026-05-07
Performance
64k
tokens
405.9
PP tok/s
13.8
TG tok/s
161470
TTFT (ms)
14.8
Peak mem (GB)
Hardware
Chip M2 Pro (19c)
Memory 32 GB
GPU Cores 19
Software
oMLX v0.3.8
macOS macOS 15.7.5
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 497.2 46.1 11.4 GB view
4k 553.7 45.2 11.8 GB view
8k 562.7 42.2 12.3 GB view
16k 539.1 37.8 12.4 GB view
32k 488.9 21.2 13.2 GB view
64k 405.9 13.8 14.8 GB current
Batching Results
Batch Size TG tok/s Speedup
46.1 1.00×
59.8 1.30×
69.1 1.50×
137.8 2.99×