May 16, 2025
Trending News

SAP HANA Cloud introduces AI vector engine for business data

  • April 8, 2024
  • 0

A new vector engine in SAP HANA Cloud enables companies to integrate large language models with specific business data for advanced database applications. SAP is making its vector

A new vector engine in SAP HANA Cloud enables companies to integrate large language models with specific business data for advanced database applications.

SAP is making its vector engine added to SAP HANA Cloud generally available. The vector engine enables companies to combine large language models (LLMs) with company-specific real-time data and business process knowledge. This integration takes place within the SAP HANA Cloud and fits into the company’s AI strategy.

The importance of the vector engine

The vector engine must take into account the limitations of LLMs. SAP thinks about the dependence on outdated training data and the lack of company-specific data and knowledge about business processes. Means Retrieval extended generation The engine can provide LLMs with all relevant data from an organization.

Key benefits and features of the vector engine include unification of multiple model data, improved search and analysis capabilities, personalized recommendations, and optimized output of LLMs. This makes it possible to create innovative applications that can communicate with users more intuitively.

Internal use

SAP HANA Cloud plays an important role within the SAP Business Technology Platform (SAP BTP) and is used internally at SAP by more than 180 different applications and services due to its multi-model capability.

The vector engine integration positions SAP HANA Cloud as the standard database for SAP’s generative AI strategy and enables customers to create advanced user experiences in collaboration with other services within SAP BTP. In addition, SAP is working on basic models specifically for SAP-related industry and process knowledge.

SAP, like other companies, wants to help customers decrypt their data with AI. A model-agnostic approach like this, which can integrate its own data without expensive and cumbersome model retraining, is relevant. Top form

Source: IT Daily

Leave a Reply

Your email address will not be published. Required fields are marked *

Exit mobile version