Search Articles — Sudonull

https://sudonull.com/nvidia-cuda-cores-accelerated-by-ai-by-38-without-human-involvement

Multi-agent Cursor system autonomously optimized 235 CUDA cores for Blackwell B200. Results, methodology, and significance for HPC developers.

https://sudonull.com/gpu-acceleration-of-the-game-of-life-cuda-triton-benchmarks

Comparison of PyTorch, CUDA, Triton for the cellular automaton "Game of Life" on A40. Results 22.5 ms, kernel code. Study optimizations for memory-bound tasks. Read the analysis.

https://sudonull.com/gradient-descent-and-backpropagation-in-neural-networks-cuda-c

Deep dive into gradient descent and backpropagation. Implementation of neural network training algorithms from scratch on CUDA/C++ for MNIST. Detailed breakdown of mathematics and code.

https://sudonull.com/rocm-is-catching-up-to-cuda-in-ai-onerocm-and-triton

AMD unified ROCm into OneROCm for all accelerators. Triton writes kernels for AMD/Nvidia. vLLM inference on par with CUDA. Six-week releases. Test the stack on Strix Halo.

https://sudonull.com/simd-optimization-of-mandelbrot-avx2-openmp-cuda

Boosting the Mandelbrot set: from 7 FPS scalar C++ to 500+ FPS on CUDA. AVX2 intrinsics, OpenMP multithreading, GPU code. Benchmarks Ryzen 5 + RTX 3050. For middle/senior developers.

Dependencies · OpenFreeEnergy/openfe-benchmarks · GitHub

https://github.com/OpenFreeEnergy/openfe-benchmarks/network/dependencies

Contribute to OpenFreeEnergy/openfe-benchmarks development by creating an account on GitHub.

Branches · OpenFreeEnergy/openfe-benchmarks · GitHub

https://github.com/OpenFreeEnergy/openfe-benchmarks/branches/active

Contribute to OpenFreeEnergy/openfe-benchmarks development by creating an account on GitHub.

openfe-benchmarks/openfe_benchmarks/data/benchmark_systems ... -...

https://github.com/OpenFreeEnergy/openfe-benchmarks/blob/main/openfe_benchmarks/data/benchmark_systems/jacs_set/p38/ligands.sdf

More file actions More file actions History 4597 lines (4257 loc) · 132 KB main openfe-benchmarks / openfe_benchmarks / data / benchmark_systems / jacs_set / p38 ...

openfe-benchmarks/openfe_benchmarks/data/data_generation at main...

https://github.com/OpenFreeEnergy/openfe-benchmarks/tree/main/openfe_benchmarks/data/data_generation

Contribute to OpenFreeEnergy/openfe-benchmarks development by creating an account on GitHub.

Projects · openfe-benchmarks · GitHub

https://github.com/OpenFreeEnergy/openfe-benchmarks/projects?query=is%3Aopen

GitHub is where people build software. More than 150 million people use GitHub to discover, fork, and contribute to over 420 million projects.

GitHub

https://github.com/OpenFreeEnergy/openfe-benchmarks/blob/76888d2c3f9d366fc7610e101e58cd5f79bb89d2/README.md

A file for each system in the benchmark set (e.g. tyk2.py). These hold\nmethods for creating the free energy network and the system components\nnecessary to calculate the benchmark.