ServiceNow and Nvidia announce Apriel Nemotron 15B AI model to accelerate agentic AI for companies

ServiceNow and Nvidia

At its Knowledge 2025 event in Las Vegas, ServiceNow announced the new AI model called Apriel Nemotron 15B in partnership with Nvidia. The new AI language model is aimed at offering more powerful AI tools to companies. The AI model is built to help AI agents handle tasks more efficiently and make better decisions in real-time.

Apriel Nemotron 15B is a large language model (LLM) developed together by ServiceNow and Nvidia. It has 15 billion parameters, which means it can understand and process complex tasks. What makes it special is that it is faster and cheaper to run, while still delivering strong performance. It’s trained using data from both companies and optimised to run on Nvidia’s cloud-based systems.

This new AI model is designed to help AI agents—software tools that assist with business tasks—work better. These agents can be used in areas like customer service, HR, and IT support. With Apriel Nemotron 15B, these tools can now understand context, follow rules, and make smart decisions without human help.

ServiceNow is also adding Nvidia’s NeMo microservices to its platform. These tools help gather and analyse business data quickly. When combined with ServiceNow’s Workflow Data Fabric, the system can learn from real-time data, helping the AI model improve over time. This feedback loop, often called a “data flywheel,” makes the AI more accurate and more personalised for users.

Apriel Nemotron 15B model will be available in the second quarter of 2025. This launch builds on previous work between the two companies, including tools to test AI agents and other AI model integrations. Their goal is to make enterprise AI tools that are not just powerful, but also safe and reliable to use.