Pinned Loading

1. vllm-project/vllm vllm-project/vllm Public

   A high-throughput and memory-efficient inference and serving engine for LLMs

   Python 60.3k 10.6k

2. DBGroup-SUSTech/GPU-Merkle-Patricia-Trie DBGroup-SUSTech/GPU-Merkle-Patricia-Trie Public

   An optimized Merkle Patricia Trie implementation on GPU, fully compatible with and integrable into Ethereum. The paper is published on VLDB 2024.

   Cuda 14

3. DBGroup-SUSTech/GPUDB-Prefetch DBGroup-SUSTech/GPUDB-Prefetch Public

   Source code of our DaMoN@SIGMOD 2024 paper "How Does Software Prefetching Work on GPU Query Processing?"

   Cuda 9

4. AlayaDB-AI/ParaGraph AlayaDB-AI/ParaGraph Public

   A cross-modal vector index with fast construction on heterogeneous CPU-GPU environment. Published on DaMoN@SIGMOD 2025.

   Cuda 13

5. wsl2000/NeuroTTT wsl2000/NeuroTTT Public

   Brain-to-text with test time training

   Python 23