ServiceNow’s message this quarter was crystal clear and served as another wake-up call for Indian IT firms making flashy announcements about AI. For ServiceNow, like Accenture, and to some extent Cognizant, AI is not an experiment anymore. It is the core of their business.
ServiceNow reported another strong quarter, beating every topline and profit target. Q3 subscription revenue rose 20.5% year-over-year in constant currency to $3.299 billion, surpassing the high end of guidance.
Current remaining performance obligations (RPO) hit $11.35 billion, up 20.5%, while total RPO climbed 23% to $24.3 billion.
ServiceNow also raised its full-year 2025 subscription revenue guidance by $55 million, now projecting a revenue of around $12.8 billion, representing 20% annual growth. The company also raised its operating and free cash flow margin outlook, citing efficiency and demand for AI-driven products.
But the real story here isn’t just the numbers. It’s how those numbers are being powered.
ServiceNow is an AI Company
ServiceNow announced an expanded partnership with NVIDIA to unite intelligent workflows and open models to scale what it calls “trusted AI” across industries. At the centre of the collaboration is Apriel 2.0, the next generation of ServiceNow’s open model family, co-developed with NVIDIA.
The model, built on NVIDIA’s Nemotron technology and trained on joint data from both companies, is designed for faster, smaller and more cost-efficient performance.
Apriel 2.0 brings advanced reasoning and multimodal capabilities, capable of interpreting screenshots, forms and diagrams—directly into enterprise workflows. It’s meant to power both autonomous and semi-autonomous agents at scale, offering low-latency, multi-step reasoning that enterprises can deploy safely, even in regulated sectors like banking, healthcare and telecom.
“The next wave of AI is about more than innovation. It’s about execution—how fast and how responsibly enterprises can put advanced intelligence to work,” Pat Casey, ServiceNow’s CTO, said. “Our collaboration with NVIDIA is built around that idea.”
Kari Ann Briski, vice president of generative AI for enterprise at NVIDIA, said the partnership reflects a shift toward more transparent and controllable enterprise AI. “Open models give enterprises the transparency and control they need to specialise AI to their data, workflows and trust standards,” she said.
ServiceNow and NVIDIA will also collaborate to modernise data centre operations, integrating ServiceNow workflows with NVIDIA’s AI Factory for Government reference design. The goal is to make enterprise infrastructure smarter, faster and more autonomous.
This includes out-of-the-box AI agents to automate retail service requests, handle public-sector tasks like citizen and interagency requests and optimise data centre asset management. In other words, the next generation of “AI ops” won’t just monitor systems, it’ll manage them.
Payoff Continues
That strategy continues to pay off. ServiceNow’s AI products are on pace to exceed half a billion dollars in annual contract value (ACV) this year and the company is sticking to its target of $1 billion by 2026. It closed 12 Now Assist deals worth over $1 million in new ACV this quarter, including one over $10 million.
Its AI Control Tower deal volume more than quadrupled compared to the previous quarter. “Every business process in every industry is being refactored for agentic AI,” ServiceNow CEO Bill McDermott said earlier this year. Q3 only reinforced that message.
The company has already shown how fast it can turn AI hype into real money. In the last quarter, it reported 21 deals that included five or more AI products and one of its largest AI deals ever—over $20 million in value. That pace has continued.
Apriel 2.0 now extends that momentum, taking ServiceNow deeper into enterprise reasoning and multimodal AI. The model follows Apriel Nemotron 15B, launched earlier this year, and Apriel-1.5-15B-Thinker, a small open-source model capable of advanced reasoning on a single GPU. Both were designed to make enterprise-grade AI practical and cost-efficient.
Apriel 2.0 is the clearest signal yet that ServiceNow is betting on smaller, high-reasoning models as the future. Gartner’s projection that such models will surpass general-purpose LLMs by 2027 fits neatly with this direction.
For ServiceNow, it’s not about competing with massive consumer models—it’s about dominating enterprise AI use cases with faster, safer, cheaper reasoning models. The company’s AI narrative is now its growth engine. Its unified AI platform—combining workflow automation, data fabric and custom models—is giving it measurable traction where others are still piloting.
Indian IT firms, for example, continue to test AI tools but rarely break out revenue from them apart from HCLTech in this quarter. ServiceNow, meanwhile, is quantifying, productising and scaling AI across its customer base.
With its NVIDIA partnership, Apriel 2.0 launch and strong financials, ServiceNow is signalling what the next phase of enterprise AI looks like—smaller models, bigger results and no waiting around.
At this rate, the only question left is how soon ServiceNow hits the $1 billion AI milestone. It might happen faster than anyone expects.
The post $12.8 Billion Reasons Why ServiceNow Wins Enterprise AI appeared first on Analytics India Magazine.
