Customers Can Harness Their Business Data in Cloud Solutions from SAP Using Customized LLMs Deployed with NVIDIA AI Foundry Services and New NVIDIA NIM Microservices

 SAP  and NVIDIA  announced a partnership expansion focused on accelerating enterprise customers’ ability to harness the transformative power of data and generative AI across SAP’s portfolio of cloud solutions and applications.

The companies are collaborating to build and deliver SAP Business AI, including scalable, business-specific generative AI capabilities inside the Joule copilot from SAP and across SAP’s portfolio of cloud solutions and applications – all of which are underpinned by the SAP generative AI hub. The generative AI hub facilitates relevant, reliable and responsible business AI and provides instant access to a broad range of large language models (LLMs).

As part of SAP’s ongoing initiative to build generative AI directly into the applications that power the world’s businesses, the partnership aims to help customers adopt generative AI capabilities at scale across their organizations. SAP will use NVIDIA’s generative AI foundry service to fine-tune LLMs for domain-specific scenarios and deploy applications with new NVIDIA NIM™ microservices. SAP and NVIDIA plan to make the new integrated capabilities available by the end of 2024.

“Enterprise customers want to leverage state-of-the-art technology that delivers real business value,” said Christian Klein, CEO and Member of the Executive Board of SAP SE. “Strategic technology partnerships, like the one between SAP and NVIDIA, are at the core of our strategy to invest in technology that maximizes the potential and opportunity of AI for business. NVIDIA’s expertise in delivering AI capabilities at scale will help SAP accelerate the pace of transformation and better serve our customers in the cloud.”

“SAP is sitting on a gold mine of enterprise data that can be transformed into custom generative AI agents to help customers automate their businesses,” said Jensen Huang, founder and CEO of NVIDIA. “Together, NVIDIA and SAP will bring custom generative AI to the thousands of enterprises around the world that rely on SAP to power their operations.”

Harnessing Business Data and Generative AI to Advance Customer Insights

SAP and NVIDIA plan to collaborate to integrate generative AI into cloud solutions from SAP, which include the latest release of the SAP Datasphere solution, SAP Business Technology Platform (SAP BTP) and RISE with SAP.

SAP plans to build additional generative AI capabilities within SAP BTP using NVIDIA’s generative AI foundry service, featuring NVIDIA DGX™ Cloud AI supercomputing, NVIDIA AI Enterprise software and NVIDIA AI Foundation models. These new capabilities are designed to be the basis of SAP’s development and deployment of generative AI for customers and is expected to be accessible in the generative AI hub in SAP AI Core and SAP Datasphere.

Additional generative AI initiatives include:

• New capabilities for the Joule copilot: Joule can leverage retrieval-augmented generation (RAG) capabilities built by NVIDIA and SAP, which can be deployed on leading hyperscalers or SAP’s own cloud environments. As a natural-language, generative AI copilot, Joule helps customers unlock the potential in their business by automating time-consuming tasks and quickly analyzing business-critical data to deliver more intelligent, personalized experiences.
• Innovative use cases leveraging SAP S/4HANA Cloud, SAP SuccessFactors and SAP Signavio: SAP and NVIDIA are exploring more than 20 generative AI use cases where the companies can combine assets to simplify and enhance digital transformation. Among these are generative AI features that can automate enterprise resource planning with intelligent invoice matching in SAP S/4HANA Cloud; improve human resources use cases leveraging SAP SuccessFactors; and accelerate new generative AI insights from SAP Signavio to better process business recommendations and optimize SAP’s customer support processes.
• Unifying AI data sources with SAP Datasphere: Built on SAP BTP, SAP Datasphere enables integration and a unified view of semantically rich SAP data with third-party data across the enterprise landscape to help customers adapt faster to market changes and make more efficient and better informed decisions. With SAP Datasphere, customers can confidently access a high-quality data fabric using AI and machine learning (ML) models. To accelerate SAP’s federated machine learning (FedML) capabilities for SAP Datasphere, NVIDIA and SAP are facilitating easier access to data for data scientists and enhancing ML workload performance with the support of NVIDIA accelerated computing platforms and NVIDIA AI Enterprise data science software such as NVIDIA RAPIDS™.
• Using LLMs for the ABAP programming language: To aid developers in creating domain-specific language code, SAP plans to use NVIDIA AI foundry services to assist in fine-tuning LLMs. This will build on SAP’s use of generative AI models to assist developers who use ABAP through the company’s ABAP Cloud model and SAP Cloud Application Programming model.

NVIDIA AI Enterprise Powers Production-Grade Generative AI Across Cloud Solutions from SAP

Once models are ready for deployment in cloud solutions from SAP, SAP plans to use NVIDIA AI Enterprise software, including NVIDIA NIM inference microservices and NVIDIA NeMo Retriever™ microservices. NVIDIA NIM can be used to accelerate and maximize inference performance across the accelerated infrastructure from SAP. Using NVIDIA NeMo Retriever microservices, SAP plans to add RAG capabilities that enable generative AI applications to more securely access data running on SAP software to improve accuracy and insights. Customers can plan to use RAG on both SAP and third-party data.