Pinned Loading

1. delta_compression_framework delta_compression_framework Public

   A delta compression framework. Implement NTransform, Finesse, Odess, Palantir.

   C 7 5

2. warp_decode warp_decode Public

   An unofficial re-implementation of Cursor's warp decode MoE inference technique. Achieved nearly 2.0x speedup on MOE layer on Nvidia H20 at batch size 1.

   Python 2