← Back to community benchmarks

MiniMax-M3

M3 Ultra (80c) · 512 GB · 4bit · 2026-07-23

Performance

195k

tokens

198.3

PP tok/s

12.6

TG tok/s

1008441

TTFT (ms)

252.4

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.5.3

macOS macOS 26.3.1

Context 200,000

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
32k	312.1	18.3	231.9 GB	view
64k	274.1	16.8	234.9 GB	view
128k	230.5	14.4	243.5 GB	view
195k	198.3	12.6	252.4 GB	current