How Google Uses Stable Diffusion

This second installment explores how Ray’s higher-level libraries—Serve, Data, and Train—abstract the complexities of running AI workloads on Google's TPU slices. Ray Serve uses a simple topology configuration to correctly gang-schedule large multi-host models, while Ray Data eliminates data-loading bottlenecks by feeding accelerators directly with native JAX batches. Finally, JaxTrainer streamlines distributed training across TPUs by automatically handling cross-slice coordination, checkpointing, and fault tolerance.

The article introduces Keras Hub, a unified library for pretrained models that simplifies access to both natural language processing (NLP) and computer vision (CV) architectures.

The article discusses optimizing Large Language Model (LLM) inference at the edge using TensorFlow Lite (TFLite) and XNNPack.

The article announces the availability of an early, experimental on-device text-to-image generation solution for Android developers using MediaPipe.

The article highlights the achievements and activities of Google Machine Learning communities in the second quarter of 2023, showcasing various training campaigns, community events, and innovative ...