Create, Migrate, and Optimize Your AI Models with
Intel® Gaudi® AI Accelerators
Discover the resources, guidance, tools, and support needed to more easily and flexibly build new AI models, migrate existing ones, and optimize model performance to meet your requirements. Access the latest Intel® Gaudi® software to build or update your infrastructure.
Start Exploring
Explore the options below to access Intel Gaudi accelerators, along with model references, libraries, containers, and tools for training and deploying GenAI and LLM.
Connect to Intel® Developer Cloud for Intel Gaudi 2 AI accelerators or Amazon EC2* DL1 for first-gen Intel Gaudi accelerators.
Find detailed instructions and videos to get started with GPU migration, working with Hugging Face models, and new customer onboarding.
Optimize, fine-tune, debug, and profile your model to meet your performance targets.
Review training and inference model performance data on the Intel Gaudi AI accelerator.
Intel® Gaudi® 3
The third generation AI accelerator delivers competitive performance for generative AI (GenAI) enterprise applications.
Developer Fundamentals for Intel Gaudi AI Accelerators
Join our expert-led workshop series to polish your skills and gain hands-on experience with cutting-edge AI technologies.
Intel® Cloud.University
Get a comprehensive overview of the Intel Gaudi AI accelerator and earn a competency badge at Intel® Cloud.University.
Latest News
Read Blogs |
Watch Videos
Explore a range of current topics and informative tutorials.
Distributed Training with FSDP
Run scalable model development with fully sharded data parallel (FSDP) training using PyTorch* and Intel Gaudi accelerators.
Build a RAG Application
Use retrieval augmented generation to deploy optimized solutions on Intel Gaudi accelerators with Hugging Face*.
Introducing Intel Gaudi Software v1.17.0
Numerous enhancements and updates deliver an improved user experience.
Visual Question and Answering
Use GenAI to create a chatbot for visual question and answering based on images.