← Back to community benchmarks

gemma-4-31b-it

M3 Ultra (80c) · 512 GB · 8bit · 2026-04-16

Performance

64k

tokens

256.4

PP tok/s

7.9

TG tok/s

255606

TTFT (ms)

41.7

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.5.dev1

macOS macOS 26.3.1

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
64k	256.4	7.9	41.7 GB	current
128k	163.6	4.9	51.3 GB	view