Listen to this Post
In a major leap for enterprise AI, ServiceNow has announced a groundbreaking AI model, Apriel Nemotron 15B, developed in collaboration with NVIDIA. This new model is designed to accelerate AI capabilities across large organizations, enabling AI agents to respond in real time and handle complex workflows in IT, HR, customer service, and more. The announcement, made at ServiceNow’s Knowledge 2025 conference, marks a significant evolution in the field of AI-powered enterprise solutions.
The Rise of Apriel Nemotron 15B: A New Standard for Enterprise AI
ServiceNow’s new model, Apriel Nemotron 15B, was unveiled at the Knowledge 2025 event with NVIDIA CEO Jensen Huang joining ServiceNow CEO Bill McDermott during the keynote. This innovative model, developed with NVIDIA’s NeMo and utilizing the Llama Nemotron Post-Training Dataset, is designed to enhance reasoning capabilities within enterprise applications. It represents a leap forward in AI technology, particularly for industries relying on fast, accurate, and scalable AI-driven workflows.
At its core, Apriel Nemotron 15B is built for reasoning and inference, making it adept at understanding complex workflows, drawing conclusions, and making decisions in real time. Unlike many large-scale models that rely on trillions of parameters, this model’s compact size and optimized design allow for faster response times and lower inference costs, without sacrificing the advanced reasoning needed for enterprise applications.
The
A Closed Loop for Continuous Improvement
ServiceNow and NVIDIA are also introducing a novel data flywheel architecture, designed to continuously improve AI performance. By integrating ServiceNow’s Workflow Data Fabric with NVIDIA’s NeMo microservices, the Apriel Nemotron 15B model can refine its capabilities over time. This closed-loop system uses workflow data to personalize responses and improve accuracy, ensuring that AI agents can handle more complex tasks as they learn from each interaction.
Moreover, the architecture provides a level of control to customers, with built-in security and compliance guardrails to ensure that data is used responsibly and securely. This data-driven approach allows businesses to maintain a high level of transparency while benefiting from enhanced AI-driven workflows.
Real-World Impact: AstraZeneca and Beyond
ServiceNow is already showing how Apriel Nemotron 15B is transforming enterprise operations. During a demo, the model was showcased in use at AstraZeneca, where it helps employees resolve issues and make decisions faster, boosting productivity and reducing manual workload. In fact, this deployment is expected to free up 90,000 hours for AstraZeneca’s workforce, highlighting the real-world value of advanced AI systems.
Jon Sigler, ServiceNow’s Executive Vice President of Platform and AI, emphasized the model’s purpose-built reasoning capabilities, which make it ideal for enterprise applications. Unlike generic AI models, Apriel Nemotron 15B combines real-time data, workflow context, and advanced reasoning to deliver tangible productivity gains.
Kari Briski, NVIDIA’s Vice President of Generative AI Software, echoed this sentiment, pointing out that the model’s small size doesn’t limit its power. By leveraging both NVIDIA’s open dataset and ServiceNow’s domain-specific data, Apriel Nemotron 15B is able to deliver fast, accurate, and cost-effective AI services that support businesses worldwide.
The Future of Enterprise AI: Scaling for the New Era
The partnership between ServiceNow and NVIDIA represents a shift in how enterprise AI is approached. Rather than relying on static models, businesses can now take advantage of intelligent systems that evolve and adapt to new challenges. As enterprises face growing demands for fast, scalable AI solutions, Apriel Nemotron 15B offers a powerful tool that balances performance with cost-efficiency.
For business leaders, the introduction of AI agents powered by Apriel Nemotron 15B means faster issue resolution, improved productivity, and more responsive digital experiences. The model’s availability will be rolled out in stages following Knowledge 2025, integrated with ServiceNow’s existing AI offerings, and is expected to become a key engine for the company’s agentic AI capabilities.
What Undercode Says:
The collaboration between ServiceNow and NVIDIA marks a critical milestone in the evolution of enterprise AI. What stands out about Apriel Nemotron 15B is its ability to handle real-time reasoning at a scale that most AI models simply cannot. The focus on inference, rather than sheer parameter count, reflects a growing trend in the AI space where efficiency and precision are prioritized over brute force computational power.
For enterprises, this model opens up new opportunities for automation and intelligent decision-making, particularly in industries where workflows are complex and time-sensitive. The ability to process large amounts of data in real-time, while maintaining the flexibility to adapt to unique business requirements, could make AI agents much more integrated into day-to-day operations.
Furthermore, the introduction of the closed-loop learning system ensures that these models can continuously evolve, offering long-term value to enterprises by improving over time. The closed-loop system, supported by ServiceNow’s Workflow Data Fabric, provides an added layer of customization and control, which is crucial for organizations concerned about data security and compliance.
In terms of scalability, the model is an excellent fit for global businesses looking to deploy AI at scale without incurring the high costs typically associated with running large AI models. The lower inference costs and faster response times make it a viable solution for industries that require rapid action, like customer service or IT.
The real-world impact, as demonstrated with AstraZeneca, suggests that this new model could redefine productivity standards in enterprise AI applications. The 90,000 hours saved by employees could easily translate into millions of dollars in cost savings, making the Apriel Nemotron 15B a valuable asset for any business looking to maximize efficiency.
Fact Checker Results
Accuracy of claims: The article’s claims about the Apriel Nemotron 15B’s capabilities align with industry trends, especially regarding AI’s move from large, general-purpose models to more efficient, purpose-built systems.
Verification: The use of NVIDIA’s NeMo and ServiceNow’s domain-specific data in training the model is consistent with current AI development practices.
Real-world impact: The reported 90,000 hours saved at AstraZeneca is credible, as real-world case studies of AI implementation often show similar productivity gains.
Prediction
Looking ahead, the Apriel Nemotron 15B could pave the way for further advancements in enterprise AI. As businesses continue to demand more agile, efficient, and intelligent systems, this model’s ability to balance speed, cost, and reasoning power could set a new benchmark for AI solutions in industries like healthcare, finance, and customer service. Additionally, as the model evolves with continuous learning, it’s likely that its impact on global enterprise operations will only grow, making AI agents an indispensable part of digital transformation strategies.
References:
Reported By: blogs.nvidia.com
Extra Source Hub:
https://www.quora.com
Wikipedia
Undercode AI
Image Source:
Unsplash
Undercode AI DI v2
