← Back to community benchmarks

Qwen3.6-35B-A3B

M4 Pro (20c) · 48 GB · 4bit · 2026-04-29

Performance

4k

tokens

776.4

PP tok/s

72.9

TG tok/s

5276

TTFT (ms)

19.0

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	658.3	77.6	18.3 GB	view
4k	776.4	72.9	19.0 GB	current
8k	763.4	67.2	19.4 GB	view
16k	711.9	56.9	20.0 GB	view
32k	619.6	44.9	21.3 GB	view
64k	489.1	31.5	24.0 GB	view
128k	325.0	20.8	29.4 GB	view