← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (20c) · 64 GB · 4bit · 2026-05-14
Performance
64k
tokens
1,354
PP tok/s
34.9
TG tok/s
48415
TTFT (ms)
17.7
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.5
Context 65,536
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,509 73.3 14.3 GB view
4k 1,956 70.4 14.9 GB view
8k 2,036 65.5 15.1 GB view
16k 2,008 57.3 15.2 GB view
32k 1,784 47.1 16.0 GB view
64k 1,354 34.9 17.7 GB current
Batching Results
Batch Size TG tok/s Speedup
73.3 1.00×
107.0 1.46×
137.7 1.88×
164.8 2.25×