Intel has announced a collaboration with numerous partner companies to establish a platform for utilizing AI within organizations, enabling the easy creation of retrieval-augmented generation (RAG) apps. Intel commits to leading the way with the first codebase supporting Xeon and Gaudi chips, providing a cloud for developers to use. They will then expand support for other organizations to further develop the project for broader use.
The exact details of this platform remain unclear at the moment. However, at this year’s Intel Vision event, Intel showcased that Xeon alone can now run LLM models. While some RAG apps currently have cloud providers offering services, these apps are directly tied to the cloud. Although some allow for switching LLM models, there are already several open-source RAG apps in existence. It remains to be seen how they will differ.
TLDR: Intel partners with multiple companies to create an AI platform for developing RAG apps, starting with support for Xeon and Gaudi chips. The specifics of the platform are unclear, but Intel demonstrated the ability to run LLM models using Xeon. Existing RAG apps are linked to cloud providers, with open-source alternatives also available.
Leave a Comment