← Back to community benchmarks

Qwen3.5-27B-Claude-4.6-Opus-Distilled

M4 Pro (20c) · 48 GB · 4bit · 2026-03-10

Performance

8k

tokens

116.0

PP tok/s

13.9

TG tok/s

70640

TTFT (ms)

16.6

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 48 GB

GPU Cores 20

Software

oMLX v0.2.6

macOS macOS 15.7.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	122.4	15.3	15.8 GB	view
4k	124.2	14.8	16.1 GB	view
8k	116.0	13.9	16.6 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	15.3	1.00×
2×	25.1	1.64×
4×	18.4	1.20×