← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 4bit · 2026-05-08
Performance
195k
tokens
1,282
PP tok/s
28.8
TG tok/s
156037
TTFT (ms)
26.2
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.8
macOS macOS 26.4
Context 200,000
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,227 117.2 15.5 GB view
4k 3,333 111.6 16.2 GB view
8k 3,644 104.3 16.3 GB view
16k 3,692 94.5 16.8 GB view
32k 3,288 79.1 17.2 GB view
64k 2,566 59.5 18.9 GB view
128k 1,751 40.5 22.5 GB view
195k 1,282 28.8 26.2 GB current
Batching Results
Batch Size TG tok/s Speedup
117.2 1.00×
171.3 1.46×
235.0 2.01×
292.9 2.50×