Trending

Show HN: I Parallelized RNN Training from O(T) to O(log T) Using CUDA

Gluon: a GPU programming language based on the same compiler stack as Triton

Accelerated Game of Life with CUDA / Triton

Show HN: An ncurses CUDA-based fluid simulation

Writing Speed-of-Light Flash Attention for 5090 in CUDA C++

Show HN: CUDA Fractal Renderer

Multiplatform Matrix Multiplication Kernels

Apple's MLX adding CUDA support

FP8 is ~100 tflops faster when the kernel name has "cutlass" in it

speed optimizations for Qwen Next on CUDA have been merged into llama.cpp

Learning CUDA for free on Google Colab

Optimizing Token Generation in llama.cpp's CUDA Backend

Qwen3-Next 80B-A3B llama.cpp implementation with CUDA support half-working already (up to 40k context only), also Instruct GGUFs

GPU 101 and Triton kernels

Fp8 runs ~100 tflops faster when the kernel name has "cutlass" in it

Fast CUDA DFloat11 decoding kernel

Single-File Qwen3 Inference in Pure CUDA C

[P] Sub-millisecond GPU Task Queue: Optimized CUDA Kernels for Small-Batch ML Inference on GTX 1650.

NVIDIA

Save item

cutlass

CUDA Templates and Python DSLs for High-Performance Linear Algebra

NVIDIA

Save item

cuda-samples

Samples for CUDA Developers which demonstrates features in CUDA Toolkit

NVIDIA

Save item

cuda-python

CUDA Python: Performance meets Productivity