← Back to community benchmarks

gemma-4-31B-it

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

Performance

32k

tokens

298.7

PP tok/s

11.4

TG tok/s

109711

TTFT (ms)

37.0

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.8rc1

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
16k	320.3	14.1	34.7 GB	view
32k	298.7	11.4	37.0 GB	current
64k	256.7	8.2	41.8 GB	view