Popular repositories Loading

1. cutlass cutlass Public

   Forked from NVIDIA/cutlass

   CUDA Templates and Python DSLs for High-Performance Linear Algebra

   C++ 1

2. sglang sglang Public

   Forked from sgl-project/sglang

   SGLang is a high-performance serving framework for large language models and multimodal models.

   Python 1

3. FlashKDA FlashKDA Public

   Forked from MoonshotAI/FlashKDA

   FlashKDA: high-performance Kimi Delta Attention kernels

   Cuda

4. verl verl Public

   Forked from verl-project/verl

   verl/HybridFlow: A Flexible and Efficient RL Post-Training Framework

   Python

5. verl-omni verl-omni Public

   Forked from verl-project/verl-omni

   Multimodal RL training framework for diffusion & omni models

   Python

6. Megatron-LM Megatron-LM Public

   Forked from NVIDIA/Megatron-LM

   Ongoing research training transformer models at scale

   Python