NVIDIA Launches Family of Open Reasoning AI Models for Developers and Enterprises to Build Agentic AI Platforms
NVIDIA has unveiled its new Llama Nemotron family of open reasoning AI models, designed to enhance AI agents' capabilities for complex tasks. The models, post-trained by NVIDIA, show up to 20% improved accuracy compared to base models and 5x faster inference speed than other leading open reasoning models.
Available in three sizes - Nano, Super, and Ultra - these models are optimized for different deployment needs and are accessible as NVIDIA NIM™ microservices. Major tech companies including Microsoft, SAP, ServiceNow, Accenture, and others are already integrating these models into their platforms.
The announcement includes new tools within the NVIDIA AI Enterprise software platform, such as the AI-Q Blueprint, AI Data Platform, and enhanced NIM microservices. The Nano and Super models are currently available through build.nvidia.com and Hugging Face, with free access for NVIDIA Developer Program members for development purposes.
NVIDIA ha svelato la sua nuova famiglia di modelli AI di ragionamento aperto Llama Nemotron, progettati per migliorare le capacità degli agenti AI in compiti complessi. I modelli, addestrati successivamente da NVIDIA, mostrano un miglioramento dell'accuratezza fino al 20% rispetto ai modelli base e una velocità di inferenza 5 volte superiore rispetto ad altri modelli di ragionamento aperto leader.
Disponibili in tre dimensioni - Nano, Super e Ultra - questi modelli sono ottimizzati per diverse esigenze di distribuzione e sono accessibili come microservizi NVIDIA NIM™. Grandi aziende tecnologiche tra cui Microsoft, SAP, ServiceNow, Accenture e altre stanno già integrando questi modelli nelle loro piattaforme.
L'annuncio include nuovi strumenti all'interno della piattaforma software NVIDIA AI Enterprise, come l'AI-Q Blueprint, la AI Data Platform e microservizi NIM potenziati. I modelli Nano e Super sono attualmente disponibili su build.nvidia.com e Hugging Face, con accesso gratuito per i membri del Programma Sviluppatori NVIDIA per scopi di sviluppo.
NVIDIA ha presentado su nueva familia de modelos de IA de razonamiento abierto Llama Nemotron, diseñados para mejorar las capacidades de los agentes de IA en tareas complejas. Los modelos, entrenados posteriormente por NVIDIA, muestran una mejora de precisión de hasta el 20% en comparación con los modelos base y una velocidad de inferencia 5 veces más rápida que otros modelos de razonamiento abierto líderes.
Disponibles en tres tamaños - Nano, Super y Ultra - estos modelos están optimizados para diferentes necesidades de implementación y son accesibles como microservicios NVIDIA NIM™. Grandes empresas tecnológicas, incluyendo a Microsoft, SAP, ServiceNow, Accenture y otras, ya están integrando estos modelos en sus plataformas.
El anuncio incluye nuevas herramientas dentro de la plataforma de software NVIDIA AI Enterprise, como el AI-Q Blueprint, la AI Data Platform y microservicios NIM mejorados. Los modelos Nano y Super están actualmente disponibles a través de build.nvidia.com y Hugging Face, con acceso gratuito para los miembros del Programa de Desarrolladores de NVIDIA con fines de desarrollo.
NVIDIA는 복잡한 작업을 위한 AI 에이전트의 능력을 향상시키기 위해 설계된 새로운 오픈 추론 AI 모델 Llama Nemotron 패밀리를 공개했습니다. NVIDIA에 의해 후속 훈련된 이 모델들은 기본 모델에 비해 최대 20% 향상된 정확도를 보여주며, 다른 주요 오픈 추론 모델보다 5배 빠른 추론 속도를 자랑합니다.
이 모델은 Nano, Super, Ultra의 세 가지 크기로 제공되며, 다양한 배포 요구에 최적화되어 있으며 NVIDIA NIM™ 마이크로서비스로 접근 가능합니다. Microsoft, SAP, ServiceNow, Accenture 등 주요 기술 기업들이 이미 이 모델을 그들의 플랫폼에 통합하고 있습니다.
이번 발표에는 NVIDIA AI Enterprise 소프트웨어 플랫폼 내의 새로운 도구, AI-Q Blueprint, AI Data Platform 및 향상된 NIM 마이크로서비스가 포함되어 있습니다. Nano 및 Super 모델은 현재 build.nvidia.com 및 Hugging Face를 통해 제공되며, NVIDIA 개발자 프로그램 회원에게는 개발 목적으로 무료로 접근할 수 있습니다.
NVIDIA a dévoilé sa nouvelle famille de modèles d'IA à raisonnement ouvert Llama Nemotron, conçue pour améliorer les capacités des agents IA dans des tâches complexes. Les modèles, formés par NVIDIA, montrent une précision améliorée allant jusqu'à 20 % par rapport aux modèles de base et une vitesse d'inférence 5 fois plus rapide que d'autres modèles de raisonnement ouvert leaders.
Disponibles en trois tailles - Nano, Super et Ultra - ces modèles sont optimisés pour différents besoins de déploiement et sont accessibles en tant que microservices NVIDIA NIM™. De grandes entreprises technologiques, dont Microsoft, SAP, ServiceNow, Accenture et d'autres, intègrent déjà ces modèles dans leurs plateformes.
L'annonce comprend de nouveaux outils au sein de la plateforme logicielle NVIDIA AI Enterprise, tels que l'AI-Q Blueprint, l'AI Data Platform et des microservices NIM améliorés. Les modèles Nano et Super sont actuellement disponibles via build.nvidia.com et Hugging Face, avec un accès gratuit pour les membres du Programme Développeurs NVIDIA à des fins de développement.
NVIDIA hat seine neue Familie von offenen Denkmodellen Llama Nemotron vorgestellt, die darauf ausgelegt sind, die Fähigkeiten von KI-Agenten bei komplexen Aufgaben zu verbessern. Die Modelle, die von NVIDIA nachtrainiert wurden, zeigen eine bis zu 20% verbesserte Genauigkeit im Vergleich zu Basis-Modellen und eine 5-mal schnellere Inferenzgeschwindigkeit als andere führende offene Denkmodelle.
Verfügbar in drei Größen - Nano, Super und Ultra - sind diese Modelle für unterschiedliche Bereitstellungsbedürfnisse optimiert und als NVIDIA NIM™ Mikrodienste zugänglich. Große Technologieunternehmen wie Microsoft, SAP, ServiceNow, Accenture und andere integrieren diese Modelle bereits in ihre Plattformen.
Die Ankündigung umfasst neue Werkzeuge innerhalb der NVIDIA AI Enterprise Softwareplattform, wie das AI-Q Blueprint, die AI Data Platform und verbesserte NIM Mikrodienste. Die Nano- und Super-Modelle sind derzeit über build.nvidia.com und Hugging Face verfügbar, mit kostenlosem Zugang für Mitglieder des NVIDIA Entwicklerprogramms zu Entwicklungszwecken.
- 20% improvement in model accuracy compared to base models
- 5x faster inference speed than competing open reasoning models
- Partnership with major tech companies for widespread adoption
- Reduced operational costs for enterprises through improved inference performance
- Ultra model not yet available for immediate release
- Production use requires paid NVIDIA AI Enterprise subscription
Insights
NVIDIA's launch of the Llama Nemotron reasoning models represents a strategic expansion in the company's AI product portfolio, specifically targeting the high-growth agentic AI segment. This announcement is significant for three key reasons:
First, the technical differentiation is substantial. The post-training enhancements deliver
Second, the partnership ecosystem demonstrates exceptional market validation. The collaboration with enterprise heavyweights like Microsoft, Accenture, SAP, ServiceNow, and others suggests strong market pull rather than technology push. This validates both the technology and market demand.
Third, NVIDIA's platform strategy is evident in how these models integrate with their broader AI Enterprise software platform, NIM microservices, and other tools. This creates multiple reinforcing revenue drivers and strengthens the company's competitive moat.
The announcement also reflects NVIDIA's evolving business model beyond hardware. By offering these models through their AI Enterprise software platform, NVIDIA is expanding recurring revenue opportunities while creating complementary demand for their GPU infrastructure.
The availability of these models through both hosted APIs and enterprise deployments gives NVIDIA flexibility in capturing value across different customer segments and deployment preferences, further strengthening their market position in enterprise AI infrastructure.
The Llama Nemotron family represents a significant technical achievement in the reasoning AI space that cements NVIDIA's position beyond hardware acceleration into the full AI stack.
What's technically notable is NVIDIA's post-training approach. Rather than building models from scratch, they've leveraged the open Llama architecture and applied specialized post-training techniques to optimize for reasoning-specific tasks. This demonstrates efficient use of compute resources while achieving substantial performance gains, suggesting NVIDIA has developed proprietary methods for model enhancement that don't require complete retraining.
The tiered deployment strategy with Nano, Super, and Ultra variants shows sophisticated product differentiation that addresses diverse computational environments from edge to multi-GPU servers. This flexibility is important for enterprise adoption where deployment scenarios vary widely.
The integration of these models as NIM microservices is particularly important as it standardizes deployment interfaces, making integration into existing systems more straightforward. The connection to NVIDIA's broader AI Enterprise platform creates a cohesive technical ecosystem rather than isolated point solutions.
The emphasis on providing the tools and techniques used to develop these models is strategically significant. By open-sourcing these elements, NVIDIA enables customization while maintaining their position in the value chain through the hardware and software infrastructure required to implement these solutions at scale.
The AI-Q Blueprint and AgentIQ toolkit reinforce NVIDIA's commitment to agentic systems, providing critical infrastructure components for the development of collaborative AI systems that can reason and act autonomously.
- Post-Trained by NVIDIA, New Llama Nemotron Reasoning Models Provide Business-Ready Foundation for Agentic AI
- Accenture, Amdocs, Atlassian, Box, Cadence, CrowdStrike, Deloitte, IQVIA, Microsoft, SAP and ServiceNow Pioneering Reasoning AI Agents With NVIDIA to Transform Work
SAN JOSE, Calif., March 18, 2025 (GLOBE NEWSWIRE) -- GTC -- NVIDIA today announced the open Llama Nemotron family of models with reasoning capabilities, designed to provide developers and enterprises a business-ready foundation for creating advanced AI agents that can work independently or as connected teams to solve complex tasks.
Built on Llama models, the NVIDIA Llama Nemotron reasoning family delivers on-demand AI reasoning capabilities. NVIDIA enhanced the new reasoning model family during post-training to improve multistep math, coding, reasoning and complex decision-making.
This refinement process boosts accuracy of the models by up to
Leading agent AI platform pioneers — including Accenture, Amdocs, Atlassian, Box, Cadence, CrowdStrike, Deloitte, IQVIA, Microsoft, SAP and ServiceNow — are collaborating with NVIDIA on its new reasoning models and software.
“Reasoning and agentic AI adoption is incredible,” said Jensen Huang, founder and CEO of NVIDIA. “NVIDIA’s open reasoning models, software and tools give developers and enterprises everywhere the building blocks to create an accelerated agentic AI workforce.”
NVIDIA Post-Training Boosts Accuracy and Reliability for Enterprise Reasoning
Built to deliver production-ready AI reasoning, the Llama Nemotron model family is available as NVIDIA NIM™ microservices in Nano, Super and Ultra sizes — each optimized for different deployment needs.
The Nano model delivers the highest accuracy on PCs and edge devices, the Super model offers the best accuracy and highest throughput on a single GPU, and the Ultra model will provide maximum agentic accuracy on multi-GPU servers.
NVIDIA conducted extensive post-training on NVIDIA DGX™ Cloud using high-quality curated synthetic data generated by NVIDIA Nemotron™ and other open models, as well as additional curated datasets cocreated by NVIDIA.
The tools, datasets and post-training optimization techniques used to develop the models will be openly available, giving enterprises the flexibility to build their own custom reasoning models.
Agentic Platforms Team With NVIDIA to Enhance Reasoning for Industries
Agentic AI platform industry leaders are working with the Llama Nemotron reasoning models to deliver advanced reasoning to enterprises.
Microsoft is integrating Llama Nemotron reasoning models and NIM microservices into Microsoft Azure AI Foundry. This expands the Azure AI Foundry model catalog with options for customers to enhance services like Azure AI Agent Service for Microsoft 365.
SAP is tapping Llama Nemotron models to advance SAP Business AI solutions and Joule, the AI copilot from SAP. Additionally, it is using NVIDIA NIM and NVIDIA NeMo™ microservices to promote increased code completion accuracy for SAP ABAP programming language models.
“We are collaborating with NVIDIA to integrate Llama Nemotron reasoning models into Joule to enhance our AI agents, making them more intuitive, accurate and cost effective,” said Walter Sun, global head of AI at SAP. “These advanced reasoning models will refine and rewrite user queries, enabling our AI to better understand inquiries and deliver smarter, more efficient AI-powered experiences that drive business innovation.”
ServiceNow is harnessing Llama Nemotron models to build AI agents that offer greater performance and accuracy to enhance enterprise productivity across industries.
Accenture has made NVIDIA Llama Nemotron reasoning models available on its AI Refinery platform — including new industry agent solutions announced today — to enable clients to rapidly develop and deploy custom AI agents tailored to industry-specific challenges, accelerating business transformation.
Deloitte is planning to incorporate Llama Nemotron reasoning models into its recently announced Zora AI agentic AI platform designed to support and emulate human decision-making and action with agents that include deep functional- and industry-specific business knowledge and built-in transparency.
NVIDIA AI Enterprise Delivers Essential Tools for Agentic AI
Developers can deploy NVIDIA Llama Nemotron reasoning models with new NVIDIA agentic AI tools and software to streamline the adoption of advanced reasoning in collaborative AI systems.
All part of the NVIDIA AI Enterprise software platform, the latest agentic AI building blocks include:
- The NVIDIA AI-Q Blueprint, which enables enterprises to connect knowledge to AI agents that can autonomously perceive, reason and act. Built with NVIDIA NIM microservices, the blueprint integrates NVIDIA NeMo Retriever™ for multimodal information retrieval and enables agent and data connections, optimization and transparency using the open-source NVIDIA AgentIQ toolkit.
- The NVIDIA AI Data Platform, a customizable reference design for a new class of enterprise infrastructure with AI query agents built with the AI-Q Blueprint.
- New NVIDIA NIM microservices, which optimize inference for complex agentic AI applications and enable continuous learning and real-time adaptation across any environment. The microservices ensure reliable deployment of the latest models from leading model builders including Meta, Microsoft and Mistral AI.
- NVIDIA NeMo microservices, which provide an efficient, enterprise-grade solution to quickly establish and maintain a robust data flywheel that enables AI agents to continuously learn from human- and AI-generated feedback. The NVIDIA AI Blueprint for building a data flywheel will offer a reference architecture for developers to easily build and optimize data flywheels using NVIDIA microservices.
Availability
The NVIDIA Llama Nemotron Nano and Super models and NIM microservices are available as a hosted application programming interface from build.nvidia.com and Hugging Face. Access for development, testing and research is free for members of the NVIDIA Developer Program.
Enterprises can run Llama Nemotron NIM microservices in production with NVIDIA AI Enterprise on accelerated data center and cloud infrastructure. Developers can sign up to be notified when NVIDIA NeMo microservices are publicly available.
The NVIDIA AI-Q Blueprint is expected to be available in April. The NVIDIA AgentIQ toolkit is available now on GitHub.
About NVIDIA
NVIDIA (NASDAQ: NVDA) is the world leader in accelerated computing.
For further information, contact:
Anna Kiachian
NVIDIA Corporation
+1-650-224-9820
akiachian@nvidia.com
Certain statements in this press release including, but not limited to, statements as to: the benefits, impact, availability, and performance of NVIDIA’s products, services, and technologies; third parties adopting NVIDIA’s products and technologies and the benefits and impact thereof; NVIDIA’s open reasoning models, software and tools giving developers and enterprises everywhere the building blocks to create an accelerated agentic AI workforce are forward-looking statements that are subject to risks and uncertainties that could cause results to be materially different than expectations. Important factors that could cause actual results to differ materially include: global economic conditions; our reliance on third parties to manufacture, assemble, package and test our products; the impact of technological development and competition; development of new products and technologies or enhancements to our existing product and technologies; market acceptance of our products or our partners' products; design, manufacturing or software defects; changes in consumer preferences or demands; changes in industry standards and interfaces; unexpected loss of performance of our products or technologies when integrated into systems; as well as other factors detailed from time to time in the most recent reports NVIDIA files with the Securities and Exchange Commission, or SEC, including, but not limited to, its annual report on Form 10-K and quarterly reports on Form 10-Q. Copies of reports filed with the SEC are posted on the company's website and are available from NVIDIA without charge. These forward-looking statements are not guarantees of future performance and speak only as of the date hereof, and, except as required by law, NVIDIA disclaims any obligation to update these forward-looking statements to reflect future events or circumstances.
Many of the products and features described herein remain in various stages and will be offered on a when-and-if-available basis. The statements above are not intended to be, and should not be interpreted as a commitment, promise, or legal obligation, and the development, release, and timing of any features or functionalities described for our products is subject to change and remains at the sole discretion of NVIDIA. NVIDIA will have no liability for failure to deliver or delay in the delivery of any of the products, features or functions set forth herein.
© 2025 NVIDIA Corporation. All rights reserved. NVIDIA, the NVIDIA logo, DGX, NVIDIA NeMo, NVIDIA Nemotron, NVIDIA NeMo Retriever and NVIDIA NIM are trademarks and/or registered trademarks of NVIDIA Corporation in the U.S. and other countries. Other company and product names may be trademarks of the respective companies with which they are associated. Features, pricing, availability and specifications are subject to change without notice.
A photo accompanying this announcement is available at https://www.globenewswire.com/NewsRoom/AttachmentNg/6b111210-07b7-4296-83fa-8c18c9acfbfc
FAQ
What performance improvements do NVIDIA's new Llama Nemotron models offer?
What are the different sizes of Llama Nemotron models available for NVDA?
Which major companies are adopting NVIDIA's new Llama Nemotron models?
How can developers access NVIDIA's new Llama Nemotron models?
