Best of GPUMay 2025

  1. 1
    Article
    Avatar of itnextITNEXT·52w

    AI: Introduction to Ollama for local LLM launch

    Ollama provides an easy way to run large language models locally on your own hardware. The guide covers installation on Linux, setting up GPU acceleration with NVIDIA cards, basic commands for model management, and integration with Python applications. It demonstrates running DeepSeek-R1 models, monitoring performance metrics, adjusting context windows, and creating custom models using Modelfiles with system prompts. Local deployment offers cost savings, privacy benefits, and the ability to experiment with models not available through public APIs.

  2. 2
    Article
    Avatar of chromeChrome Developers·1y

    What's New in WebGPU (Chrome 137)

    Chrome 137 introduces several updates and features to WebGPU, including the use of GPUTextureView for externalTexture binding, allowing buffer copies without offsets and sizes, and enhancements to WGSL workgroupUniformLoad. It also highlights changes in GPUAdapterInfo attributes and the removal of some experimental options. Dawn updates improve WebGPU development with new bindings and compatibility improvements.

  3. 3
    Article
    Avatar of hnHacker News·1y

    VictorTaelin/WebMonkeys: Massively parallel GPU programming on JavaScript, simple and clean.

    WebMonkeys allows massively parallel GPU programming in JavaScript through a simple API. It supports browser and Node.js environments, enabling tasks like array processing and mining operations using the GPU. Users can define libraries, handle raw data buffers, and employ efficient parallel algorithms without the complexity typically associated with WebGL.

  4. 4
    Article
    Avatar of watercoolerWatercooler·1y

    Abstractions