← Back to community benchmarks

gemma-4-26B-A4B-it

M3 Ultra (60c) · 256 GB · 4bit · 2026-05-19
Performance
16k
tokens
1,684
PP tok/s
84.7
TG tok/s
9731
TTFT (ms)
15.8
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.9.dev1
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,290 107.0 14.3 GB view
4k 1,677 102.1 15.0 GB view
8k 1,720 93.2 15.3 GB view
16k 1,684 84.7 15.8 GB current
32k 1,519 70.6 16.7 GB view
64k 1,238 51.0 18.7 GB view
Batching Results
Batch Size TG tok/s Speedup
107.0 1.00×
151.0 1.41×
210.6 1.97×
279.8 2.61×