← Back to community benchmarks

Ornith-1.0-35B-OptiQ

M3 Ultra (60c) · 256 GB · 4bit · 2026-07-21

Performance

32k

tokens

1,632

PP tok/s

65.8

TG tok/s

20085

TTFT (ms)

23.1

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.5.1

macOS macOS 26.5.2

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	2,151	78.2	21.7 GB	view
4k	2,220	77.5	22.5 GB	view
8k	2,173	73.7	22.6 GB	view
16k	1,936	70.5	22.7 GB	view
32k	1,632	65.8	23.1 GB	current
64k	1,197	59.2	23.8 GB	view