← Back to community benchmarks

Qwen3.5-35B-A3B-5bit

M1 Pro (16c) · 32 GB · 5bit · 2026-03-14

Performance

4k

tokens

286.9

PP tok/s

37.8

TG tok/s

14279

TTFT (ms)

24.2

Peak mem (GB)

Hardware

Chip M1 Pro (16c)

Memory 32 GB

GPU Cores 16

Software

oMLX v0.2.11

macOS macOS 26.2

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	269.4	40.8	24.1 GB	view
4k	286.9	37.8	24.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	40.8	1.00×
2×	50.2	1.23×
4×	56.1	1.38×