← Back to community benchmarks

gemma-4-26B-A4B-it

M3 Ultra (60c) · 256 GB · 8bit · 2026-05-22
Performance
16k
tokens
1,645
PP tok/s
71.0
TG tok/s
9959
TTFT (ms)
27.1
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.9
macOS macOS 26.4.1
Context 16,384
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 1,342 84.6 25.8 GB view
4k 1,657 81.0 26.4 GB view
8k 1,706 78.0 26.6 GB view
16k 1,645 71.0 27.1 GB current
32k 1,484 60.8 28.2 GB view
64k 1,215 46.3 30.1 GB view