← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Ultra (80c) · 512 GB · 4bit · 2026-05-10

Performance

4k

tokens

2,488

PP tok/s

89.1

TG tok/s

1646

TTFT (ms)

20.0

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.3.9.dev1

macOS macOS 15.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,663	91.5	19.2 GB	view
4k	2,488	89.1	20.0 GB	current
32k	2,112	76.1	22.4 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	91.5	1.00×
2×	156.9	1.71×
4×	260.2	2.84×