← Back to community benchmarks

Qwen3.5-35B-A3B

M1 Ultra (64c) · 128 GB · 4bit · 2026-03-12

Performance

4k

tokens

1,059

PP tok/s

56.8

TG tok/s

3869

TTFT (ms)

20.2

Peak mem (GB)

Hardware

Chip M1 Ultra (64c)

Memory 128 GB

GPU Cores 64

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,001	58.8	20.0 GB	view
4k	1,059	56.8	20.2 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	58.8	1.00×
2×	104.5	1.78×
4×	162.3	2.76×