21.7 C
New York

Nvidia launches NIM to simplify AI model deployment

Published:

During the GTC conference, Nvidia introduced NIM, a new software platform aimed at facilitating the integration of custom and pre-trained AI models into production settings.

NIM, part of Nvidia’s latest announcements at the GTC conference, leverages Nvidia’s expertise in AI model inferencing and optimization, providing developers with a more straightforward approach. By combining AI models with an optimized inferencing engine and packaging them into accessible containers as microservices, NIM significantly reduces deployment time. This streamlined process, according to TechCrunch, can now be swiftly completed, eliminating the need for extensive in-house AI knowledge.

The platform supports models from well-known entities like NVIDIA, A121, and Getty Images, as well as open models from tech leaders such as Google and Meta. Through collaborations with Amazon, Google, and Microsoft, Nvidia aims to integrate NIM microservices into major cloud platforms to broaden accessibility for developers.

NIM’s Foundation: Nvidia’s Inferencing Engines

At the core of NIM are the Triton Inference Server, TensorRT, and TensorRT-LLM, showcasing Nvidia’s dedication to offering a strong base for AI applications. Additionally, the platform includes specialized microservices like Riva for speech and translation adjustments, cuOpt for routing optimizations, and the Earth-2 model for weather and climate simulations.

Manuvir Das, Nvidia’s head of enterprise computing, highlighted NIM’s efficiency and high-quality for enterprises. NIM allows developers to concentrate on building enterprise applications without the burden of managing models.

NIM epitomizes Nvidia’s vision of transforming enterprises into AI-driven entities by providing a suite of containerized AI microservices. With support from industry leaders and a network of partners, Nvidia’s NIM is set to revolutionize how AI models are deployed and used across various sectors.

Jensen Huang, Nvidia’s CEO, emphasized the transformative potential of NIM, envisioning a future where every enterprise utilizes AI to enhance operations and foster innovation.

Related articles

Recent articles