← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Max (40c) · 128 GB · 4bit · 2026-04-29

Performance

16k

tokens

2,724

PP tok/s

93.9

TG tok/s

6015

TTFT (ms)

21.0

Peak mem (GB)

Hardware

Chip M3 Max (40c)

Memory 128 GB

GPU Cores 40

Software

oMLX v0.3.7rc2

macOS macOS 26.4.1

Context 16,384

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,182	104.8	18.7 GB	view
4k	1,456	99.6	19.4 GB	view
16k	2,724	93.9	21.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	104.8	1.00×
2×	167.6	1.60×
4×	239.4	2.28×