← Back to community benchmarks

gemma-4-26b-a4b-it

M4 Pro (20c) · 64 GB · 8bit · 2026-05-22
Performance
32k
tokens
573.8
PP tok/s
35.6
TG tok/s
57106
TTFT (ms)
27.8
Peak mem (GB)
Hardware
Chip M4 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.9
macOS macOS 26.3
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 670.0 45.2 25.8 GB view
4k 721.7 43.8 26.4 GB view
8k 720.8 42.6 26.6 GB view
32k 573.8 35.6 27.8 GB current
64k 472.3 27.1 30.0 GB view
Batching Results
Batch Size TG tok/s Speedup
45.2 1.00×
67.0 1.48×
84.4 1.87×