← Back to community benchmarks

carl-omni9b-v15-vision

M3 Ultra (60c) · 96 GB · 4bit · 2026-04-29

Performance

32k

tokens

1,021

PP tok/s

74.1

TG tok/s

32088

TTFT (ms)

9.1

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	931.2	231.9	8.2 GB	view
4k	788.7	106.6	6.5 GB	view
8k	1,108	101.5	6.9 GB	view
16k	1,086	90.9	7.6 GB	view
32k	1,021	74.1	9.1 GB	current