← Back to community benchmarks

Qwen3.6-35B-A3B-UD

M1 Ultra (64c) · 128 GB · 4bit · 2026-05-18

Performance

4k

tokens

535.3

PP tok/s

62.0

TG tok/s

7651

TTFT (ms)

21.2

Peak mem (GB)

Hardware

Chip M1 Ultra (64c)

Memory 128 GB

GPU Cores 64

Software

oMLX v0.3.8

macOS macOS 26.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	890.9	192.2	21.7 GB	view
4k	535.3	62.0	21.2 GB	current
8k	1,119	61.1	21.5 GB	view