Skip to content
View zhouwg's full-sized avatar
  • just for fun
  • China
  • 03:22 (UTC +08:00)

Organizations

@ggml-hexagon

Block or report zhouwg

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
zhouwg/README.md

Hi there 👋

I'm a Linux veteran who built my own agentic Linux distribution with agents, based on Hyprland + QuickShell + llama.cpp. I work with a Dell Vostro 5890 (equipped with Ubuntu 26.04 + 11th Gen Intel(R) Core(TM) i7-11700F (16) @ 4.90 GHz CPU + 64 GB memory + 256 GB SSD + 2 TB HDD ), a Dell Vostro 3267 (equipped with Pop!_OS 24.04) , other x86 laptops, and Snapdragon-powered Android phones (8 Gen3 & 8 Gen4). More testing and hands-on practical work is needed to turn this promising platform into something people can easily install and use. This is a work in progress. I'm resolving compatibility issues one machine at a time: starting with QEMU, then physical x86-64 hardware, followed by physical aarch64 hardware.

My next planned device is the Gorgon Halo (AMD Ryzen AI Max+ PRO 495 mini workstation, 16C/32T Zen5 up to 5.2GHz, Radeon 8065S 40CU RDNA3.5, XDNA2 NPU 55 TOPS, 192GB onboard LPDDR5X-8533 unified memory, 2TB SSD). I don’t use any Apple ecosystem products. My preference is Linux-based platforms, including Android.

My technical background covers Linux OS internals, Android OS internals, embedded systems, streaming media, and virtualization. As a long-time Linux programmer and a newcomer to the AI field, I am the original author and maintainer of FastRPC-based ggml-hexagon. I have submitted contributions to llama.cpp that have not been accepted upstream.

I like .vimrc and I can’t lie --- I don’t lie in most situations, though sometimes I do(adult life is not easy, and sometimes you have to lie).

Pinned Loading

  1. ggml-hexagon/ggml-hexagon ggml-hexagon/ggml-hexagon Public

    Forked from ggml-org/llama.cpp

    the original FastRPC-based implementation of a specified llama.cpp backend for Qualcomm Hexagon NPU, history of ggml-hexagon: https://github.com/ggml-hexagon/ggml-hexagon/discussions/18

    C++ 56 9

  2. try-omarchy-linux try-omarchy-linux Public

    Run/Try Omarchy on Linux with minimal setup

    C

  3. kantv kantv Public

    workbench for learning and practicing on-device AI technologies under real-world scenarios on Android smartphones with online TV. powered by llama.cpp + whisper.cpp + FFmpeg + opencv-mobile

    Java 200 27

  4. popos_iso popos_iso Public

    Forked from pop-os/iso

    Pop!_OS ISO production

    Makefile