Home / 𧱠AI Foundation Stack / ①Inference / Serving
chiennv2000/orthrus
Fast, lossless LLM inference via dual-view diffusion decoding.
Commercial license
β Commercial OKMIT
ε―εη¨οΌιεΈΈεͺιδΏηθδ½ζ¬θ²ζ/ζζ¬ζ’ζ¬Ύ
Topics
diffusion-language-modelsefficient-inferencelarge-language-modelsllmllm-efficiencymodel-architecturenatural-language-processing
Ad