search
NVIDIA CUDA
Trends
- 1Janus tool runs local AI models on any GPU via Vulkan●Show HN: Janus – Go binary that runs GGUF models via Vulkan on AMD/Intel/Nvidia
A new open-source project called Janus, released on GitHub by Vibra-Ingenn, lets users run GGUF-format AI models locally through a single Go binary using Vulkan graphics drivers. It works across AMD, Intel and Nvidia GPUs, bypassing the usual need for CUDA or vendor-specific toolchains. Developers on Hacker News are discussing its simplicity and cross-vendor GPU support.
- 2Hobbyist reverse-engineers NVIDIA CUDA binaries with AI clustering●Motivation I am not a data scientist/engineer, but I wanted to do a little experiment on... # nvidia # ai # clustering #
A self-described non-data-scientist has run a small experiment applying data clustering and AI techniques to reverse-engineer NVIDIA's compiled CUDA binaries. The write-up walks through the motivation and method, showing how machine learning can help make sense of opaque proprietary code without a formal engineering background. The project is drawing attention from the developer and AI community as an example of accessible, do-it-yourself analysis.
Repos
- ymcrcat/rgpu Keep Python on your laptop. Run PyTorch operations and hold tensors on a remote GPU, including from a Mac with no CUDA i
- magnitudedev/magnitude Open source inference engine for agents that optimizes itself for your exact hardware. Compiles and tunes its kernels on
- tile-ai/tilelang Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels