← Back to community benchmarks

Laguna-S-2.1-oQ4e-fast

M3 Ultra (60c) · 96 GB · 4bit · 2026-07-25

Performance

8k

tokens

656.5

PP tok/s

55.5

TG tok/s

12479

TTFT (ms)

60.4

Peak mem (GB)

Hardware

Chip M3 Ultra (60c)

Memory 96 GB

GPU Cores 60

Software

oMLX v0.5.3

macOS macOS 15.7.7

Context 8,192

Performance by Context Length

Context	PP tok/s	TG tok/s	Peak Mem
1k	678.7	58.6	59.8 GB	view
4k	678.0	57.4	60.1 GB	view
8k	656.5	55.5	60.4 GB	current
16k	625.8	53.5	61.0 GB	view
32k	565.8	47.9	62.1 GB	view
64k	475.8	41.6	64.5 GB	view
128k	360.6	33.8	69.2 GB	view