deepseek-ai/DeepJIT

DeepJIT is a lightweight, header-only C++20 JIT runtime for NVIDIA CUDA GPUs and HUAWEI Ascend (昇腾) NPUs. It gives C++/Python extension authors a shared interface for compiling kernel source at runtime, caching the resulting binaries, loading them onto the device, and launching them with backend-specific options. DeepJIT handles the JIT infrastructure so that kernel libraries can focus on their device code. Both backends share runtime configuration, source and include hashing, in-memory and on-disk caches, and lazy initialization. Kernel source and compiler/launch options remain specific to the selected backend. Main authors: @guyan364, @kurisu6912, @LyricZhao. The host environment must provide Linux, a C++20 compiler and standard library with std::format support, and the dependencies for the selected backend. The default GIL support also requires Python and pybind11; consumers that manage the GIL themselves can disable it as described under Integration. DeepJIT is intended to be embedded into your extension as a header-only dependency.…