← Back to community benchmarks

Qwen3.5-35B-A3B

M3 Ultra (80c) · 512 GB · bf16 · 2026-03-12

Performance

4k

tokens

2,101

PP tok/s

54.8

TG tok/s

1949

TTFT (ms)

66.6

Peak mem (GB)

Hardware

Chip M3 Ultra (80c)

Memory 512 GB

GPU Cores 80

Software

oMLX v0.2.9

macOS macOS 15.4.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,352	55.2	66.5 GB	view
4k	2,101	54.8	66.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	55.2	1.00×
2×	74.5	1.35×
4×	129.8	2.35×