← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Ultra (80c) · 96 GB · 4bit · 2026-04-24

Performance

4k

tokens

2,550

PP tok/s

94.0

TG tok/s

1606

TTFT (ms)

20.0

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 96 GB

GPU Cores 80

Software

oMLX v0.3.7

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,531	96.5	19.2 GB	view
4k	2,550	94.0	20.0 GB	current
8k	2,684	88.5	20.4 GB	view
16k	2,553	80.1	21.0 GB	view
32k	2,207	63.3	22.4 GB	view