← Back to community benchmarks

Llama-3.2-1B-Instruct

M4 (8c) · 16 GB · 4bit · 2026-03-12

Performance

4k

tokens

1,060

PP tok/s

77.3

TG tok/s

3867

TTFT (ms)

1.4

Peak mem (GB)

Hardware

Chip M4 (8c)

Memory 16 GB

GPU Cores 8

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 4,096

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,122	107.5	1.3 GB	view
4k	1,060	77.3	1.4 GB	current

Batching Results

Batch Size	TG tok/s	Speedup
1×	107.5	1.00×
2×	181.0	1.68×
4×	183.0	1.70×