← Back to community benchmarks

DeepSeek-R1-Distill-Qwen-32B

M2 Max (38c) · 96 GB · 4bit · 2026-03-11

Performance

4k

tokens

155.0

PP tok/s

16.2

TG tok/s

26434

TTFT (ms)

19.0

Peak mem (GB)

Hardware

Chip M2 Max (38c)

Memory 96 GB

GPU Cores 38

Software

oMLX v0.2.6

macOS macOS 26.2

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	157.6	17.5	17.9 GB	view
4k	155.0	16.2	19.0 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	17.5	1.00×
2×	29.3	1.67×
4×	32.0	1.83×