← Back to community benchmarks

gemma-4-26b-a4b-it

M5 Max (40c) · 128 GB · 4bit · 2026-04-03
Performance
32k
tokens
2,546
PP tok/s
32.6
TG tok/s
12869
TTFT (ms)
16.9
Peak mem (GB)
Hardware
Chip M5 Max (40c)
Memory 128 GB
GPU Cores 40
Software
oMLX v0.3.2
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 2,255 80.9 15.1 GB view
4k 3,382 69.5 15.5 GB view
8k 3,461 61.5 15.6 GB view
16k 3,007 44.9 16.1 GB view
32k 2,546 32.6 16.9 GB current
64k 1,664 20.9 18.6 GB view
Batching Results
Batch Size TG tok/s Speedup
80.9 1.00×
106.9 1.32×
133.2 1.65×
157.1 1.94×