reims-vgpu is an experimental virtual GPU for macOS guests that uses QEMU to decode the guest's GPU command stream and execute it through Metal or Vulkan, leveraging the built-in AppleParavirtGPU driver without requiring custom kexts.

Stars

143

7-day growth

No data

Forks

7

Open issues

3

License

LGPL-3.0

Last updated

2026-07-27

AI repository intelligence
FR-AI / ANALYSIS

Why it is worth attention

It enables accelerated graphics in macOS VMs without modifying the guest OS, by implementing the device side of Apple's paravirtual GPU interface and translating commands via Metal or Vulkan, which is a novel approach in the open-source virtualization space.

Who it is for

  • Developers working on GPU virtualization for macOS
  • QEMU contributors interested in GPU device models
  • Researchers exploring Metal-to-Vulkan translation
  • Hobbyists experimenting with macOS VMs on non-Apple hardware

Use cases

  • Running macOS 13 Ventura guests with hardware-accelerated graphics on Linux or Apple Silicon hosts
  • Testing and debugging Metal/Vulkan translation via metal2vulkan
  • Developing and validating custom GPU command stream decoders
  • Evaluating snapshot-revert workflows for reproducible VM testing

Strengths

  • No custom guest driver or kext installation required—uses existing AppleParavirtGPU.kext
  • Supports three distinct host/guest/backend pathways (x86+Linux+Vulkan, arm64+macOS+Metal, arm64+macOS+Vulkan)
  • Provides snapshot-revert boot scripts that protect golden images from corruption
  • Open source under LGPL-3.0-or-later with a modular Rust codebase and vendored QEMU shims

Considerations

  • Alpha stage—ABI, behavior, and supported pathways may change without notice
  • Requires significant host configuration (KVM, Vulkan/Metal stacks, custom QEMU build)
  • Not production-ready; focused on experimentation and bring-up

README quick start

Getting started

Description

reims-vgpu is an experimental virtual GPU for macOS guests

Related repositories

Similar projects matched by category, topics, and programming language.

l0ng-ai
Featured
l0ng-ai GitHub avatar

tty7

tty7 is a fast, GPU-accelerated terminal workbench with persistent sessions, built-in input tools, SSH support, and coding agent awareness.

Developer ToolsCLI & Terminal
359
Blaizzy
Featured
Blaizzy GitHub avatar

nativ

Nativ is a native macOS app that lets you run AI models locally on Apple silicon, offering chat, model management, performance analytics, and an OpenAI/Anthropic-compatible API server.

AI & Machine LearningAI Vision & Audio
985
makecindy
Featured
makecindy GitHub avatar

cindy

Cindy is an open-source AI agent that runs locally on your machine, integrates multiple AI harnesses and models, and provides memory, skills, and automation to perform real work in your projects and apps.

AI & Machine LearningLarge Language Models
958