Introduction to Gen AI at SAP
Generative AI has the potential to significantly impact businesses in today’s world. It refers to AI systems that can not only analyze data and make predictions, but also create new content, designs, and solutions across a wide range of applications, from automating repetitive tasks to creating personalized customer experiences. GenAI can help streamline workflows, improve decision-making, and drive innovation. It can also be used to generate new product designs, marketing materials, and even virtual prototypes.
In addition, GenAI can help businesses better understand their customers through advanced data analysis and predictive modeling. This can lead to more targeted marketing campaigns, improved customer service, and ultimately, increased customer satisfaction and loyalty.
This workshop will introduce the Generative AI hub and the SAP AI Launchpad.
SAP AI Core and the generative AI hub help you to integrate LLMs and AI into new business processes in a cost-efficient manner.
Large Language Models
Large Language Models, or LLMs, are self-supervised, deep learning models that have been trained on vast amounts of unlabeled data. They leverage AI technology and industrial-scale computational resources to learn complex language patterns and semantic knowledge bases for natural language processing (NLP) tasks. They parse input, such as prompts, and return contextually relevant responses written in natural language. A single LLM can perform multiple NLP tasks by using different input formats and output modes.
The generative AI hub gives instant access to a broad range of LLMs from different providers, such as Google, OpenAI, Amazon and many others that are both proprietary or open-source. With this access, it’s possible to orchestrate multiple scenarios, whether programmatically via an SDK or directly through the SAP AI Launchpad.
LLMs are general models but can be fine-tuned with additional embeddings for specialized or domain-specific use cases.
Model Access
Access to generative AI models is provided under the global AI scenario foundation-models, which is managed by SAP AI Core. Individual models are provided as executables in the form of serving templates, and accessed by choosing the corresponding template for the desired model.
For more information about the available models in the SAP Generative AI Hub, including conversion rates for tokens, rate limits and deprecation dates, see SAP Note 3437766
Orchestration
The orchestration service runs on SAP AI Core under the global AI scenario orchestration. It provides unified access to multiple generative AI models through consistent code, configuration, and deployment.
Orchestration offers a harmonized API that allows you to use different foundation models without changing the client code. To use different foundation models and versions, you need to create at least one orchestration deployment, or use the orchestration deployment in your default resource group.
Key features of orchestration include:
Templating: This feature lets you compose prompts with placeholders that are populated during inference.
Content filtering: Allows you to restrict the type of content passed to and received from generative AI models.
Data masking: Enables anonymization or pseudonymization of data before passing it to a generative AI model. With pseudonymization, masked data in the model’s response is automatically restored.
Grounding: This feature lets you integrate external, domain-specific, or real-time data to enhance pretrained models with contextually relevant information beyond their general training material.
Translation: Enables the addition of translation capabilities for both input and output in your orchestration workflow.
The focus of this Technical Academy workshop is to show how Generative AI can be used to assist companies with a typical business process, and how this can be even further enhanced by employing the use of the above features inclusing Retrieval Augmented Generation (RAG) in conjunction with SAP HANA Cloud’s vector engine to create more contextually relevant responses to queries.
For more information on the Generative AI Hub and SAP AI Core in general, please visit the AI Core Landing Page on the SAP help portal.