← Back to community benchmarks

DeepSeek-R1-Distill-Qwen-32B

M4 Pro (16c) · 48 GB · 4bit · 2026-04-18

Performance

8k

tokens

80.6

PP tok/s

10.7

TG tok/s

101665

TTFT (ms)

19.9

Peak mem (GB)

Hardware

Chip M4 Pro (16c)

Memory 48 GB

GPU Cores 16

Software

oMLX v0.3.6

macOS macOS 26.4.1

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	88.3	12.7	17.9 GB	view
4k	86.6	11.8	19.0 GB	view
8k	80.6	10.7	19.9 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	12.7	1.00×
2×	25.2	1.98×
4×	50.7	3.99×