← Back to community benchmarks

gemma-4-31B-it-oQ4

M3 Ultra (80c) · 512 GB · 4bit · 2026-05-13
Performance
32k
tokens
294.4
PP tok/s
22.3
TG tok/s
111288
TTFT (ms)
25.4
Peak mem (GB)
Hardware
Chip M3 Ultra (80c)
Memory 512 GB
GPU Cores 80
Software
oMLX v0.3.9.dev2
macOS macOS 26.4
Context 32,768
Performance by Context Length
Context PP tok/s TG tok/s Peak Mem
1k 326.8 29.5 18.1 GB view
4k 334.1 28.5 20.7 GB view
8k 332.5 27.2 21.3 GB view
16k 318.2 25.2 22.4 GB view
32k 294.4 22.3 25.4 GB current
64k 241.3 17.2 31.3 GB view