← Back to community benchmarks

MiniMaxM3-IQ4

M3 Ultra (80c) · 512 GB · 4bit · 2026-07-17

Performance

16k

tokens

317.7

PP tok/s

17.1

TG tok/s

51570

TTFT (ms)

221.7

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	348.6	24.9	218.6 GB	view
4k	354.9	17.8	219.7 GB	view
8k	329.0	17.6	220.1 GB	view
16k	317.7	17.1	221.7 GB	current
32k	300.7	16.6	224.8 GB	view
64k	282.8	15.5	231.5 GB	view
128k	258.4	13.3	244.4 GB	view