Migrating Figma's rendering backend from WebGL to WebGPU required modernizing interfaces, eliminating global GPU state, and enabled better performance, GPU parallelism, and clearer error handling.
RAPIDS enables zero-code GPU acceleration for pandas, scikit-learn, NetworkX, and other Python data libraries, delivering large speedups and scalable GPU-native workflows.
GPULlama3.java Brings GPU-Accelerated LLM Inference to Pure Java
The TornadoVM programming guide demonstrates how developers can utilize hardware-agnostic APIs, enabling the same Java source code to run identically on various hardware accelerators.