← Back to community benchmarks

gemma-4-31b-it-oQ8

M3 Ultra (80c) · 512 GB · 8bit · 2026-05-01

Performance

16k

tokens

306.1

PP tok/s

15.7

TG tok/s

53527

TTFT (ms)

36.1

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	318.0	18.2	31.8 GB	view
4k	329.4	16.9	34.5 GB	view
8k	314.6	17.0	35.0 GB	view
16k	306.1	15.7	36.1 GB	current
32k	289.6	14.0	39.0 GB	view
64k	240.6	12.6	45.1 GB	view
128k	183.9	8.9	57.1 GB	view
195k	137.7	7.0	72.1 GB	view