ServiceNow has entered a new era of enterprise AI with the unveiling of Apriel Nemotron 15B, a robust reasoning model developed in collaboration with NVIDIA. This model aims to enhance AI's ability to handle complex tasks, streamline workflows, and provide real-time support in functions such as IT, HR, and customer service. The announcement was made at ServiceNow’s Knowledge 2025 conference, where NVIDIA CEO and founder Jensen Huang joined ServiceNow chairman and CEO Bill McDermott on stage. Smaller Model, Bigger Impact Apriel Nemotron 15B boasts 15 billion parameters, placing it in the mid-range compared to larger, general-purpose models that can exceed trillions of parameters. Despite its relatively compact size, this model is highly efficient and cost-effective, delivering rapid responses and reducing inference costs. These attributes make it particularly suitable for enterprises looking to harness AI without the overhead associated with massive models. The development of Apriel Nemotron 15B leveraged NVIDIA NeMo, an open-source framework, along with ServiceNow’s domain-specific data and NVIDIA’s Llama Nemotron Post-Training Dataset. Training was conducted on NVIDIA DGX Cloud running on Amazon Web Services (AWS), which provided the necessary high-performance computing resources to fine-tune the model effectively. A Closed Loop for Continuous Learning ServiceNow and NVIDIA are introducing a novel data flywheel architecture that integrates ServiceNow’s Workflow Data Fabric with NVIDIA NeMo microservices. This system allows for continuous learning and improvement of AI performance by personalizing responses based on workflow data. Importantly, guardrails are in place to ensure that data usage is secure and compliant with customer regulations. The key components of this architecture include NeMo Customizer and NeMo Evaluator, which facilitate the refinement and assessment of AI responses. By continuously updating and adapting, Apriel Nemotron 15B can better align with the specific needs of each enterprise, leading to more accurate and contextually relevant interactions. Real-World Applications During the keynote, ServiceNow demonstrated how Apriel Nemotron 15B can be deployed in practical scenarios. One notable example is its partnership with pharmaceutical giant AstraZeneca, where AI agents will assist in resolving employee issues and facilitating decision-making with greater speed and precision. This deployment is expected to save AstraZeneca approximately 90,000 hours annually, allowing employees to focus on higher-value tasks. Jon Sigler, ServiceNow’s executive vice president of Platform and AI, emphasized the unique strengths of Apriel Nemotron 15B. He highlighted that the model’s purpose-built reasoning capabilities enable it to drive productivity in ways that generic models cannot. Sigler noted that combining real-time enterprise data, workflow context, and advanced reasoning results in more effective AI assistants. Kari Briski, vice president of generative AI software at NVIDIA, echoed Sigler’s sentiment. She stated that by optimizing for both performance and cost, Apriel Nemotron 15B sets a new standard for enterprise-ready AI. Briski mentioned that the model’s smaller footprint makes it more efficient to run, ensuring that it remains cost-effective even as it scales to meet growing business demands. Shaping the Future of Enterprise AI The partnership between ServiceNow and NVIDIA represents a significant shift in how enterprises approach AI. Traditionally, AI models were static and required substantial ongoing maintenance. Apriel Nemotron 15B and the accompanying data flywheel architecture offer a more dynamic solution, where AI systems can learn and adapt continually. This advancement is not just a win for individual businesses but also a milestone for the broader tech industry. It demonstrates the potential for AI to transform operations and productivity in various sectors, from healthcare to finance. By providing faster resolution times, improved efficiency, and more responsive digital experiences, Apriel Nemotron 15B is poised to revolutionize the way enterprises interact with and utilize AI. Availability and Next Steps ServiceNow plans to roll out AI Agents powered by Apriel Nemotron 15B following the Knowledge 2025 conference. These AI services will be integrated into ServiceNow’s Now LLM offerings and will serve as a critical component of the company’s agentic AI solutions. Companies can look forward to enhanced AI capabilities that reduce downtime and increase operational efficiency, ultimately driving business growth. According to industry experts, the introduction of Apriel Nemotron 15B signifies a strategic move by ServiceNow toStay ahead in the competitive landscape of enterprise AI. It underscores the company's commitment to innovation and its ability to adapt to the evolving needs of its clients. NVIDIA, known for its leadership in AI and graphics processing, continues to strengthen its position in the enterprise market through strategic collaborations like this one. ServiceNow, founded in 2004 and headquartered in Santa Clara, California, is a leading provider of cloud-based services for managing enterprise operations. NVIDIA, founded in 1993 and based in the same city, is at the forefront of AI and GPU technology. Together, they are poised to redefine what is possible with enterprise AI, setting a new benchmark for intelligence, efficiency, and scalability.