Pinned Loading

1. tile-ai/tilelang tile-ai/tilelang Public

   Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

   Python 9k 982

2. pytorch/torchtitan pytorch/torchtitan Public

   A PyTorch native platform for training generative AI models

   Python 5.9k 1.1k

3. uccl-project/uccl uccl-project/uccl Public

   UCCL is an efficient communication library for GPUs, covering collectives, P2P (e.g., KV cache transfer, RL weight transfer), and EP (e.g., GPU-driven)

   C++ 1.6k 198

4. NVIDIA/Megatron-LM NVIDIA/Megatron-LM Public

   Ongoing research training transformer models at scale

   Python 18.4k 4.7k

5. olo-dot-io/Uni-CLI olo-dot-io/Uni-CLI Public

   One command for every interface—search, run, and inspect real software across APIs, browsers, desktops, local tools, and MCP.

   TypeScript 228 1

6. fla-org/flash-linear-attention fla-org/flash-linear-attention Public

   🚀 Efficient implementations for emerging model architectures

   Python 6k 812