Here's an overview of the features we intend to work on in the near future.

Core Keras

Performance and Optimization

• A new Keras Pruning API to help users create smaller and more efficient models.
• Introduce comprehensive support for model quantization, including:
  • Post-training quantization techniques like GPTQ and AWQ.
  • Quantization-Aware Training (QAT) int8 support.

Scale and Distribution

• Distributed Training
  • Comprehensive guides for multi-host TPU and multi-host GPU training.
  • Official performance benchmarks
  • A Backup and Restore callback to handle preemptions gracefully during long training runs.

Integrations and Ecosystem

• Add support for exporting models to the ODML LiteRT format, simplifying deployment on edge and mobile devices.
• Integrate Qwix, a new JAX-based library for quantization.
• [Contributions Welcome] Integrate PyGrain for creating efficient, large-scale data loading and preprocessing pipelines.

Guides and Tutorials

• Deployment Guides: End-to-end tutorials on deploying Keras models to Vertex AI, and on-device via LiteRT.
• Guide on efficient inference using KerasHub models with vLLM.
• AI Agents and RAG: Advanced examples of building AI agents with function calling and creating Retrieval-Augmented Generation (RAG) pipelines.
• Training Techniques: Guides on model distillation, handling training preemptions on TPUs, and best practices for image augmentation (e.g., CutMix and MixUp).
• Others: Orbax checkpointing, FLUX model guide/example, etc.

KerasHub

See the roadmap here.

KerasRS

See the roadmap here.