← Back to community benchmarks

gemma-4-31b-it

M3 Ultra (60c) · 256 GB · 8bit · 2026-05-08
Performance
8k
tokens
251.0
PP tok/s
17.9
TG tok/s
32640
TTFT (ms)
35.0
Peak mem (GB)
Hardware
Chip M3 Ultra (60c)
Memory 256 GB
GPU Cores 60
Software
oMLX v0.3.8
macOS macOS 26.4.1
Context 8,192
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
4k 172.5 18.5 33.6 GB view
8k 251.0 17.9 35.0 GB current
16k 242.9 16.8 36.1 GB view
32k 226.7 15.2 39.2 GB view
64k 197.8 12.8 45.1 GB view
128k 153.8 9.7 57.1 GB view