Google is set to release Chrome 113 stable next month, which will finally support WebGPU. This next-generation graphics API has been under development since 2017, providing developers with lower-level access to advanced computational processing on the GPU. Google has limited this release to ChromeOS, macOS, and Windows, with more platforms like Linux to follow later this year.
WebGPU provides direct hardware access through the GPUAdapter function, allowing commands to be queued via an elaborate render pipeline. This can access each device and run complicated shaders, rendering complex visuals directly in the browser. The API is more similar to Direct3D 12, Vulkan, and Metal than WebGL, which is based on OpenGL. (ES).
This API claims to cut the JavaScript effort for the same visuals by over three times while increasing the machine-learning model interface. WebGPU gives access to more extensive GPU functionality than the WebGL API family, providing developers with first-class support for standard GPU operations.
WebGPU’s initial version will be the foundation for future upgrades and expansions. In WGSL, the WebGPU Shading Language, Google’s Chrome team intends to enable deep access to shader cores for even more machine learning optimisations and improved ergonomics.
Developers may read more about WebGPU support in Chrome on the Chrome Developers blog. They may discover W3C specs, best practices, and MDN documentation here to help them comprehend this new API.
