← Back to community benchmarks

Qwen3.5-35B-A3B

M4 Pro (16c) · 48 GB · 4bit · 2026-03-09

Performance

32k

tokens

567.9

PP tok/s

39.2

TG tok/s

57698

TTFT (ms)

21.6

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.2.6

macOS macOS 26.2

Context 32,768

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	695.1	81.6	20.1 GB	view
4k	741.3	77.8	20.2 GB	view
32k	567.9	39.2	21.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	81.6	1.00×
2×	123.3	1.51×
4×	139.5	1.71×