← Back to community benchmarks

Qwen3.6-35B-A3BoQ8

M3 Ultra (60c) · 256 GB · 8bit · 2026-04-30

Performance

8k

tokens

2,083

PP tok/s

74.6

TG tok/s

3933

TTFT (ms)

36.5

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 256 GB

GPU Cores 60

Software

oMLX v0.3.8

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,744	318.0	36.7 GB	view
4k	869.2	77.2	36.2 GB	view
8k	2,083	74.6	36.5 GB	current
16k	1,984	69.6	37.1 GB	view
32k	1,719	63.9	38.5 GB	view
64k	1,310	56.3	41.2 GB	view