← Back to community benchmarks

Qwen3.5-35B-A3B-Claude-4.6-Opus-Reasoning-Distilled

M5 Pro (20c) · 64 GB · 6bit · 2026-03-15

Performance

8k

tokens

1,880

PP tok/s

69.0

TG tok/s

4357

TTFT (ms)

27.6

Peak mem (GB)

Hardware

Chip M5 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.13

macOS macOS 26.3.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,236	77.7	27.3 GB	view
4k	1,745	74.9	27.4 GB	view
8k	1,880	69.0	27.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	77.7	1.00×
2×	115.3	1.48×
4×	150.7	1.94×