← Back to community benchmarks

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

M4 (10c) · 32 GB · 3bit · 2026-05-07
Performance
16k
tokens
305.4
PP tok/s
30.3
TG tok/s
53651
TTFT (ms)
12.7
Peak mem (GB)
Hardware
Chip M4 (10c)
Memory 32 GB
GPU Cores 10
Software
oMLX v0.3.8
macOS macOS 26.3.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 359.0 38.0 11.4 GB view
4k 360.7 36.0 12.1 GB view
8k 341.3 33.1 12.2 GB view
16k 305.4 30.3 12.7 GB current
32k 283.4 25.3 13.6 GB view
64k 232.9 16.7 14.8 GB view
Batching Results
Batch Size TG tok/s Speedup
38.0 1.00×
73.3 1.93×