← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Pro (20c) · 64 GB · 4bit · 2026-05-15
Performance
8k
tokens
2,121
PP tok/s
68.0
TG tok/s
3863
TTFT (ms)
15.3
Peak mem (GB)
Hardware
Chip M5 Pro (20c)
Memory 64 GB
GPU Cores 20
Software
oMLX v0.3.8
macOS macOS 26.5
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,522 74.8 14.3 GB view
4k 2,003 72.2 14.8 GB view
8k 2,121 68.0 15.3 GB current
16k 2,066 60.7 15.6 GB view
32k 1,807 50.4 16.5 GB view
64k 1,385 37.9 18.5 GB view
Batching Results
Batch Size TG tok/s Speedup
74.8 1.00×
103.2 1.38×
137.9 1.84×
154.1 2.06×