boostr 0.1.0

ML framework built on numr - attention, quantization, model architectures
Documentation
1
2
3
4
5
pub mod flash;
pub mod fused_qkv;
pub mod mla;
pub mod paged_attention;
pub mod varlen_attention;