← Back to community benchmarks

Qwen3.5-35B-A3B

M3 Ultra (60c) · 96 GB · 4bit · 2026-03-10

Performance

4k

tokens

1,928

PP tok/s

95.4

TG tok/s

2124

TTFT (ms)

19.4

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.2.6

macOS macOS 15.3

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,357	99.8	19.2 GB	view
4k	1,928	95.4	19.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	99.8	1.00×
2×	168.7	1.69×
4×	258.2	2.59×