← Back to community benchmarks

Qwen3.5-35B-A3B

M5 Pro (20c) · 64 GB · 4bit · 2026-03-23

Performance

1k

tokens

1,641

PP tok/s

99.3

TG tok/s

624

TTFT (ms)

20.1

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.20rc1

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,641	99.3	20.1 GB	current
4k	2,330	94.4	21.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	99.3	1.00×
2×	154.7	1.56×
4×	199.1	2.01×