wgpu4k is an open-source organization dedicated to building foundational libraries and tools around WebGPU, the next-generation graphics and compute API for the web and native platforms. Our mission is to simplify GPU programming, unify access to WebGPU implementations (such as wgpu and Dawn), and empower developers to harness the full potential of modern GPUs—whether for graphics, machine learning, scientific computing, or beyond.
WebGPU is a low-level, high-performance API designed for modern GPUs. It enables:
- Cross-platform compatibility: Run the same code on web browsers, desktop, and mobile devices.
- Performance: Direct access to GPU features for graphics and compute workloads.
- Future-proofing: Built to evolve with hardware and industry standards.
At wgpu4k, we aim to:
- Democratize GPU programming: Provide intuitive, high-level APIs that abstract the complexity of WebGPU.
- Unify the ecosystem: Bridge the gap between WebGPU implementations (e.g.,
wgpu,Dawn) and platforms (JVM, Android, iOS, WASM). - Enable diverse use cases: From real-time rendering to GPU-accelerated machine learning, our libraries are designed to be versatile and powerful.
- Foster collaboration: Build a community-driven ecosystem where developers, researchers, and engine maintainers can contribute and innovate.
| Project | Description | Platforms |
|---|---|---|
| wgpu4k | High-level, Kotlin-friendly WebGPU API for graphics and compute. Ideal for application developers. | JVM, Android, iOS, WASM, Native |
| wgpu4k-native | Low-level Kotlin wrapper for wgpu-native. Designed for engine maintainers and advanced users. | JVM (FFM), Android (JNI), iOS (C interop) |
- Cross-platform engines: Integrations with game engines and frameworks (e.g., KorGE).
- Machine learning toolkits: GPU-accelerated layers and utilities for Kotlin/ML.
- Build tools: Plugins for automated binding generation (e.g., JNI, Panama).
- Game development: Build cross-platform 2D/3D engines with WebGPU as the backend.
- Interactive media: Real-time visualizations, VR/AR, and creative coding.
- Machine learning: Accelerate inference and training with WebGPU compute shaders.
- Scientific computing: Offload simulations and data processing to the GPU.
- Write once, deploy anywhere: Target desktop, mobile, and web from a single Kotlin codebase.
- Start coding: Explore our repositories and check out the examples.
- Contribute: Submit issues, suggest features, or open pull requests.
- Join the discussion: Share ideas and ask questions in GitHub Discussions.
- Integrate wgpu4k-native: Use our low-level bindings to add WebGPU support to your engine.
- Collaborate: Partner with us to improve cross-platform compatibility.
- Experiment: Leverage WebGPU for GPU-accelerated research in ML, physics, or data science.
- Publish: Share your findings and tools with the community.
WebGPU is the future of GPU programming. wgpu4k makes it accessible, portable, and powerful—so you can focus on building the next generation of applications, games, and tools.
🔗 Explore our projects → wgpu4k/wgpu4k | wgpu4k/wgpu4k-native 💬 Join the conversation → GitHub Discussions
Let’s build the future of GPU programming—together.