← Back to community benchmarks

Qwythos-9B-v2-oQ4e-VL

M1 Pro (14c) · 16 GB · 4bit · 2026-07-20

Performance

64k

tokens

153.0

PP tok/s

16.7

TG tok/s

428476

TTFT (ms)

10.0

Peak mem (GB)

Hardware

Chip M1 Pro (14c)

Memory 16 GB

GPU Cores 14

Software

oMLX v0.5.2.dev1

macOS macOS 26.5.2

Context 65,536

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	187.5	24.8	5.9 GB	view
4k	209.3	24.3	6.5 GB	view
8k	211.8	23.5	6.7 GB	view
16k	204.5	22.2	7.2 GB	view
32k	173.1	19.8	8.1 GB	view
64k	153.0	16.7	10.0 GB	current