Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
237 lines (150 loc) · 3.89 KB

CONTRIBUTING.md

File metadata and controls

237 lines (150 loc) · 3.89 KB

Contributing to Vision-Language Runtime

First, thank you for your interest in contributing to Vision-Language Runtime.
This project aims to build a fast, modular, and production-oriented runtime for multimodal vision-language systems.

We welcome contributions from developers, researchers, and engineers interested in AI systems, WebGPU acceleration, and real-time multimodal inference.

Code of Conduct

All contributors must follow respectful and professional collaboration standards.

Expected behavior:

  • Be constructive and technical
  • Respect different levels of experience
  • Focus discussions on improving the system

Harassment, hostility, or non-technical conflicts are not tolerated.

Ways to Contribute

You can contribute in several ways:

1. Bug Reports

If you find a bug:

  1. Open a GitHub Issue
  2. Provide:
    • Steps to reproduce
    • Expected behavior
    • Actual behavior
    • Browser / OS
    • Console logs

Example issue title:

Runtime crash when loading WebGPU backend on Chrome 121

2. Feature Requests

Feature proposals should include:

  • Problem being solved
  • Proposed solution
  • Technical considerations
  • Performance impact (if relevant)

3. Performance Improvements

This project strongly values performance engineering.

Examples:

  • WebGPU optimizations
  • Model inference improvements
  • Memory management
  • Rendering pipeline optimization

Always include benchmarks when possible.

4. Documentation

You can improve:

  • README clarity
  • API documentation
  • architecture explanations
  • developer onboarding

Development Setup

Clone the repository:

git clone https://github.com/deepdevjose/Vision-Language-Runtime.git
cd Vision-Language-Runtime

Install dependencies:

npm install

Run development server:

npm run dev

Build production version:

npm run build

Coding Standards

Please follow these guidelines:

General

  • Keep functions small and focused
  • Avoid unnecessary abstractions
  • Prefer readability over cleverness

JavaScript / TypeScript

  • Use const by default
  • Use descriptive variable names
  • Avoid deeply nested logic
  • Document complex sections

Example:

// Initialize runtime pipeline
const runtime = new RuntimePipeline(config)

Pull Request Process

  1. Fork the repository

  2. Create a feature branch

    git checkout -b feature/improve-webgpu-pipeline

  3. Commit clearly

    git commit -m "Improve WebGPU tensor upload performance"

  4. Push your branch

    git push origin feature/improve-webgpu-pipeline

  5. Open a Pull Request

    Include:

    • What problem it solves
    • Technical explanation
    • Screenshots (if UI related)
    • Benchmarks (if performance related)

Issue Guidelines

Before opening an issue:

  • Check existing issues
  • Provide reproducible steps
  • Keep the report technical and precise

Good issue titles:

  • Memory leak during camera stream initialization
  • WebGPU backend fails on AMD GPUs
  • Runtime state machine enters invalid state

Performance Contributions

If your contribution affects runtime performance:

Include:

  • Benchmark environment
  • Hardware used
  • Before / after metrics

Example:

Device: Apple M2
Browser: Chrome 122

Before: 23 FPS
After: 38 FPS

Security

If you discover a vulnerability, do not open a public issue immediately.

Instead, contact the maintainers privately.

Philosophy of the Project

Vision-Language Runtime is designed around:

  • Real-time multimodal interaction
  • Edge-capable AI inference
  • WebGPU acceleration
  • Minimal latency systems

Contributions should align with these goals.

Maintainer

Project maintained by:

José Manuel Cortes Cerón
Research collaborator — Xi'an Jiaotong-Liverpool University
GitHub: https://github.com/deepdevjose