← Back to community benchmarks

Qwen3.5-35B-A3B-oQ4e

M4 Pro (16c) · 64 GB · 4bit · 2026-04-06

Performance

1k

tokens

655.6

PP tok/s

70.9

TG tok/s

1562

TTFT (ms)

21.3

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 64 GB

GPU Cores 16

Software

oMLX v0.3.4

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	655.6	70.9	21.3 GB	current
4k	738.8	66.9	22.1 GB	view
128k	821.7	20.0	27.2 GB	view
195k	613.9	18.2	30.3 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	70.9	1.00×
2×	93.7	1.32×
4×	107.4	1.51×