hopper
cuDNN Frontend is NVIDIA's modern, open-source entry point to the cuDNN library and a growing collection of high-performance open-source kernels.
Minimal GPU runtime for Python - high-performance CUDA kernels, memory management, and LLM inference without heavy dependencies
A Python DSL to write Nvidia PTX for Hopper and Blackwell in JAX and PyTorch
Cross-platform FlashAttention-2 Triton implementation for Turing+ GPUs with custom configuration mode
