ServiceNow and NVIDIA Unveil Groundbreaking AI Solutions at Knowledge 2025

The Apriel Nemotron 15B model is a compact, enterprise-grade large language model (LLM) that boasts advanced reasoning capabilities. Developed collaboratively with NVIDIA, this groundbreaking model is post-trained using data from both NVIDIA and ServiceNow, allowing it to evaluate relationships, apply rules, and make informed decisions based on complex workflows. Notably, its design focuses on delivering lower latency and reduced inference costs, which enhance the overall efficiency of AI operations.

The innovative model utilizes NVIDIA NeMo, the NVIDIA Llama Nemotron Post-Training Dataset, and domain-specific data from ServiceNow, all of which were processed on NVIDIA DGX Cloud via Amazon Web Services (AWS). Early benchmarks indicate that the Apriel Nemotron 15B performs exceptionally well within its size category, making it a promising tool for scaling agentic AI workflows.

Alongside the introduction of the Apriel Nemotron 15B, ServiceNow and NVIDIA unveiled a collaborative effort to integrate select NVIDIA NeMo microservices into ServiceNow’s Workflow Data Fabric. This integration aims to create a robust data flywheel that enhances model accuracy and facilitates personalized user experiences.

The data flywheel architecture will curate and contextualize enterprise workflow data, refining the reasoning models to ensure that customers maintain control over their data in a secure and compliant manner. This innovative closed-loop learning process is designed to improve model adaptability, ultimately resulting in highly personalized, context-aware AI agents that can significantly enhance enterprise productivity.

Jon Sigler, Executive Vice President of Platform and AI at ServiceNow, expressed enthusiasm about the new model's capabilities. “With this new Apriel Nemotron 15B reasoning model, we’re powering intelligent AI agents that can make context-aware decisions, adapt to complex workflows, and deliver personalized outcomes at scale,” he stated. He further highlighted the importance of the collaboration, emphasizing the continuous learning cycle that the data flywheel enables.

Kari Briski, Vice President of Generative AI Software for Enterprise at NVIDIA, echoed these sentiments, noting that the partnership's goal is to reimagine employee productivity with AI tools that streamline workflows. “Together, we’ve built the Apriel Nemotron 15B model to serve as an enterprise-grade reasoning engine,” Briski remarked, underscoring the potential for intelligent digital agents to transform business operations.

The capabilities of the Apriel Nemotron 15B model are set to revolutionize how businesses handle complex tasks. For instance, an AI agent leveraging this new model could efficiently resolve intricate billing issues by analyzing past customer interactions and providing accurate recommendations for next steps. With each case handled, the AI agent will continue to learn and improve, leading to faster and more accurate outcomes.

The Apriel Nemotron 15B reasoning model is expected to be available in the second quarter of 2025, marking a significant milestone in the ongoing collaboration between ServiceNow and NVIDIA. This latest development not only builds on previous advancements, such as the integration of NVIDIA Llama Nemotron models with the ServiceNow AI Platform, but also sets the stage for future innovations aimed at ensuring that AI solutions are intelligent, measurable, and securely deployable in real-world environments.

As enterprise AI continues to evolve, the partnership between ServiceNow and NVIDIA stands as a testament to the transformative power of technology in navigating complexity and enhancing operational resilience. With the introduction of the Apriel Nemotron 15B and the integration of advanced data processing capabilities, businesses can look forward to a new era of AI-driven productivity and efficiency.