明度
BRIGHT
ORDL

NLDA-MK-II Quantized

Version 2.1.1-q8

active Parameters: 64M Architecture: Transformer Q8_0
<p>Q8_0 quantized variant for edge deployment. Active training and testing in progress.</p>

Capabilities

  • Low-latency inference, mobile deployment, SIMD-optimized matrix ops