← Back to community benchmarks

LFM2-24B-A2B

M3 Max (30c) · 36 GB · 4bit · 2026-03-12

Performance

1k

tokens

1,235

PP tok/s

118.1

TG tok/s

830

TTFT (ms)

13.3

Peak mem (GB)

Hardware

Chip M3 Max (30c)

Memory 36 GB

GPU Cores 30

Software

oMLX v0.2.9

macOS macOS 26.3.1

Context 1,024

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	1,235	118.1	13.3 GB	current
4k	1,436	106.1	13.3 GB	view

Batching Results

Batch Size	TG tok/s	Speedup
1×	118.1	1.00×
2×	166.4	1.41×
4×	204.5	1.73×
8×	257.3	2.18×