← Back to community benchmarks

DeepSeek-R1-Distill-Qwen-7B

M4 Pro (20c) · 64 GB · 4bit · 2026-03-26

Performance

8k

tokens

438.5

PP tok/s

45.0

TG tok/s

18683

TTFT (ms)

5.0

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.20

macOS macOS 15.7.3

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	498.9	54.1	4.8 GB	view
4k	489.2	50.0	4.9 GB	view
8k	438.5	45.0	5.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	54.1	1.00×
2×	99.4	1.84×
4×	110.6	2.04×