2 个仓库
Awesome GitHub RepositoriesDLPack Protocols
Zero-copy array exchange between GPU computing frameworks using the DLPack protocol.
Distinct from Zero-Copy: Distinct from Zero-Copy: focuses on the DLPack protocol for framework interop, not general zero-copy serialization.
Explore 2 awesome GitHub repositories matching data & databases · DLPack Protocols. Refine with filters or upvote what's useful.
Awesome DLPack Protocols GitHub Repositories
用 AI 发现最棒的仓库。我们将通过 AI 为您搜索最匹配的仓库。
- nvidia/warp
NVIDIA/warp
6,233在 GitHub 上查看Warp is a Python framework that JIT-compiles Python functions into CUDA kernels for GPU-accelerated parallel computation, with built-in automatic differentiation and multi-framework array interoperability. At its core, it provides a GPU kernel compilation system that enables writing and executing custom GPU kernels directly from Python, while supporting automatic gradient computation through those kernels for integration with machine learning pipelines. The framework also includes tile-based cooperative computing, where thread blocks partition into tiles for shared-memory and tensor-core opera
Exchanges GPU arrays between frameworks via DLPack without copying memory for seamless interop.
Pythoncudadifferentiable-programminggpu - tile-ai/tilelang
tile-ai/tilelang
5,226在 GitHub 上查看TileLang is a Python-embedded domain-specific language compiler that JIT-compiles and autotunes GPU kernels. It uses a tile-based DSL, automatic software pipelining, and parallel autotuning to generate optimized GPU kernels at runtime. It supports tensor core operations with Pythonic syntax, automatic memory management, and thread mapping. The compiler searches over tile sizes, thread counts, and scheduling policies, compiling and benchmarking candidates in parallel to find the fastest kernel. It also caches compiled binaries and tuning results to disk for reuse across sessions. TileLang inc
Wraps tensors from DLPack-compatible frameworks for direct kernel execution.
Python
