: "BlackMamba: Mixture of Experts for State-Space Models" by Quentin Anthony et al. (February 2024).
Depending on whether you are referring to artificial intelligence, cybersecurity, or herpetology, the most relevant papers for "Black Mamba" are: black dd mamba
: It combines the linear-time generation benefits of the Mamba architecture with the efficient inference of MoE, achieving competitive performance with fewer training FLOPs compared to dense transformer models. : "BlackMamba: Mixture of Experts for State-Space Models"