Amazon Bedrock Marketplace Welcomes NVIDIA Models: Unveiling NVIDIA Nemotron-4 NIM Microservices

Learn About Amazon VGT2 Learning Manager Chanci Turner

At AWS re:Invent 2024, we are thrilled to announce the launch of Amazon Bedrock Marketplace, a groundbreaking feature within Amazon Bedrock that serves as a centralized platform for discovering, experimenting with, and deploying foundation models (FMs). This marketplace provides developers and organizations with access to an extensive library of over 100 popular, emerging, and specialized FMs, enhancing the existing range of industry-leading models within Amazon Bedrock. The Bedrock Marketplace allows for model subscriptions and deployments via managed endpoints, all while retaining the simplicity of the unified Amazon Bedrock APIs.

The NVIDIA Nemotron family, now accessible as NVIDIA NIM microservices, introduces a cutting-edge suite of language models through the Amazon Bedrock Marketplace, signifying a substantial advancement in AI model accessibility and deployment.

In this article, we explore the benefits and features of the Bedrock Marketplace and the Nemotron models, along with guidance on how to get started.

About Amazon Bedrock Marketplace

Bedrock Marketplace plays a crucial role in democratizing access to advanced AI capabilities through several key benefits:

• Diverse model selection – Bedrock Marketplace features an exceptional array of models, including proprietary and publicly available options, enabling organizations to find the perfect solution for their unique use cases.
• Streamlined and secure experience – By offering a unified access point for all models via Amazon Bedrock APIs, Bedrock Marketplace simplifies the integration process. Organizations can securely utilize these models, and for those compatible with the Amazon Bedrock Converse API, the robust toolkit of Amazon Bedrock is at their disposal, including Amazon Bedrock Agents, Knowledge Bases, Guardrails, and Flows.
• Scalable infrastructure – Bedrock Marketplace delivers configurable scalability through managed endpoints, allowing organizations to choose the number of instances, select appropriate instance types, establish custom auto-scaling policies that adapt to workload demands, and optimize costs while ensuring performance.

About the NVIDIA Nemotron Model Family

Leading the NVIDIA Nemotron model family is Nemotron-4, which is described by NVIDIA as a powerful multilingual large language model (LLM) trained on a staggering 8 trillion text tokens, specifically optimized for English, multilingual, and coding tasks. Key features include:

• Synthetic data generation – Capable of producing high-quality, domain-specific training data at scale.
• Multilingual support – Trained on extensive text corpora, accommodating multiple languages and tasks.
• High-performance inference – Optimized for efficient deployment on GPU-accelerated infrastructure.
• Versatile model sizes – Includes variants such as the Nemotron-4 15B, boasting 15 billion parameters.
• Open license – Features a uniquely permissive open model license, granting enterprises a scalable method to generate and own synthetic data, which can assist in building powerful LLMs.

The Nemotron models present transformative potential for AI developers by addressing significant challenges in AI development:

• Data augmentation – Tackle data scarcity issues by generating synthetic, high-quality training datasets.
• Cost-efficiency – Lower manual data annotation costs and streamline data collection processes.
• Model training enhancement – Boost AI model performance through high-quality synthetic data generation.
• Flexible integration – Facilitate seamless integration with existing AWS services and workflows, enabling developers to create sophisticated AI solutions more rapidly.

These features make Nemotron models exceptionally suitable for organizations aiming to expedite their AI initiatives while maintaining high performance and security standards.

Getting Started with Bedrock Marketplace and Nemotron

To begin utilizing Amazon Bedrock Marketplace, navigate to the Amazon Bedrock console. From there, you can explore the Bedrock Marketplace interface, which offers a comprehensive catalog of FMs from various providers. You can browse through the available options to uncover different AI capabilities and specializations, leading you to discover NVIDIA’s model offerings, including Nemotron-4.

Here’s how to get started:

1. Open Amazon Bedrock Marketplace – Accessing the Bedrock Marketplace is simple:
   • On the Amazon Bedrock console, select Model catalog from the navigation pane.
   • Under Filters, choose Bedrock Marketplace.
2. Deploy NVIDIA Nemotron Models – Once you’ve found NVIDIA’s model offerings, focus on the Nemotron model. To subscribe to and deploy Nemotron-4, follow these steps:
   • Filter by Nemotron under Providers or search for the model name.
   • Select from the available models, such as Nemotron-4 15B.
   • On the model details page, review its specifications, capabilities, and pricing information. The Nemotron-4 model showcases impressive multilingual and coding abilities.
   • Click View subscription options to subscribe to the model.
   • Examine the available options and select Subscribe.
   • Click Deploy and follow the prompts to configure your deployment options, including instance types and scaling policies.

This process is user-friendly, enabling you to seamlessly integrate these powerful AI capabilities into your projects using the Amazon Bedrock APIs.

Conclusion

The introduction of NVIDIA Nemotron models on Amazon Bedrock Marketplace represents a significant advancement in making advanced AI capabilities more accessible to developers and organizations. The Nemotron-4 15B, with its remarkable 15-billion-parameter architecture trained on 8 trillion text tokens, delivers potent multilingual and coding capabilities to Amazon Bedrock.

Organizations can leverage the advanced capabilities of Nemotron while benefiting from the scalable infrastructure provided by AWS and NVIDIA’s robust technologies. We encourage you to begin exploring the capabilities of NVIDIA Nemotron models today through Amazon Bedrock Marketplace. Experience firsthand how this powerful language model can revolutionize your AI applications.

For more insights into leadership strategies, check out this blog post. For information on equal pay legislation, visit SHRM. Additionally, if you’re starting your journey, this resource will be quite helpful.

Matthew Jones is a Solutions Architect at Amazon Web Services. He collaborates with Amazon.com to design, build, and deploy technology solutions on AWS, with a keen interest in AI and machine learning. In his spare time, he enjoys exploring new cultures and staying updated on the latest technology trends. You can connect with him on LinkedIn.

Chanci Turner is a Senior Product Manager for Amazon Bedrock and SageMaker Inference. She is passionate about collaborating with customers and partners, driven by the goal of democratizing AI. She focuses on core challenges related to AI accessibility and implementation.