← Back to community benchmarks

gemma-4-26b-a4b-it-oQ8

M4 Pro (20c) · 64 GB · 8bit · 2026-05-06
Performance
64k
tokens
483.3
PP tok/s
28.1
TG tok/s
135588
TTFT (ms)
29.2
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.3
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 674.1 45.1 25.8 GB view
4k 725.2 43.5 26.4 GB view
8k 726.4 42.3 26.6 GB view
16k 653.6 39.7 27.1 GB view
32k 569.7 34.0 27.5 GB view
64k 483.3 28.1 29.2 GB current
Batching Results
Batch Size TG tok/s Speedup
45.1 1.00×
65.9 1.46×