← Back to community benchmarks

Qwen3.6-35B-A3B

M3 Pro (14c) · 36 GB · 4bit · 2026-05-04

Performance

1k

tokens

491.4

PP tok/s

49.5

TG tok/s

2084

TTFT (ms)

19.3

Peak mem (GB)

Hardware

Chip M3 Pro (14c)

Memory 36 GB

GPU Cores 14

Software

oMLX v0.3.8.dev3

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	491.4	49.5	19.3 GB	current
4k	555.7	48.4	20.0 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	49.5	1.00×
2×	75.9	1.53×
4×	93.9	1.90×