Optimum documentation
Overview
You are viewing main version, which requires installation from source. If you'd like
regular pip install, checkout the latest stable version (v1.24.0).
Overview
Welcome to the Optimum for Intel® Gaudi® AI Accelerator how-to guides!
These guides tackle more advanced topics and will show you how to easily get the best from HPUs. Here’s what you’ll find:
- Script adaptation: Learn how to adapt a Transformers/Diffusers script for Intel Gaudi
- Pretraining models: A guide to pretraining a model using Transformers
- Accelerating training: Discover techniques to speed up training
- Accelerating inference Learn how to optimize inference for faster execution
- Using DeepSpeed: Scale your training to handle larger models
- Multi-node training: Speed up runs with multi-node setups
- Quantization: Explore FP8 and UINT4 quantization for optimized inference