← Back to community benchmarks

Llama-3.2-3B

M4 Pro (20c) · 64 GB · unknown · 2026-03-12

Performance

1k

tokens

1,038

PP tok/s

35.4

TG tok/s

988

TTFT (ms)

6.5

Peak mem (GB)

Hardware

Chip M4 Pro (20c)

Memory 64 GB

GPU Cores 20

Software

oMLX v0.2.9

macOS macOS 15.6.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,038	35.4	6.5 GB	current
4k	1,060	32.9	6.8 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	35.4	1.00×
2×	38.0	1.07×
4×	69.8	1.97×