← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Ultra (80c) · 512 GB · 4bit · 2026-04-18

Performance

4k

tokens

918.5

PP tok/s

90.6

TG tok/s

4460

TTFT (ms)

20.4

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.6

macOS macOS 26.5

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,542	284.4	19.9 GB	view
4k	918.5	90.6	20.4 GB	current
16k	1,814	88.2	21.3 GB	view