← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 4bit · 2026-04-13
Performance
32k
tokens
3,250
PP tok/s
52.4
TG tok/s
10084
TTFT (ms)
16.0
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.4
macOS macOS 26.4.1
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,233 111.9 14.2 GB view
4k 3,345 103.0 14.7 GB view
8k 3,628 87.1 14.8 GB view
16k 3,656 72.8 15.2 GB view
32k 3,250 52.4 16.0 GB current
64k 2,530 31.4 17.7 GB view
128k 1,725 14.3 21.2 GB view
195k 1,290 10.0 24.9 GB view
Batching Results
Batch Size TG tok/s Speedup
111.9 1.00×
162.2 1.45×
231.9 2.07×
288.9 2.58×