SAN FRANCISCO--(BUSINESS WIRE)--Weights & Biases, the AI developer platform, today announced multiple new platform integrations at NVIDIA GTC, a global AI conference running from March 18 to 21. The integrations include support for the NVIDIA DGX systems, which allows customers to use Weights & Biases software to easily access NVIDIA accelerated computing resources both in the cloud and on-premises, as well as support for NVIDIA NIM and other microservices, part of the NVIDIA AI Enterprise software platform. These new platform integrations are an addition to the existing Weights & Biases integrations with other software included with NVIDIA AI Enterprise, such as the NeMo framework for large language model (LLM) development, NVIDIA MONAI for training medical imaging models for healthcare AI applications, and NVIDIA TAO Toolkit. Weights & Biases is a DGX-Ready Software partner, NVIDIA AI Enterprise partner and a member of NVIDIA Inception.
W&B Launch, part of the Weights & Biases AI developer platform, provides seamless portability of AI workloads across compute clusters, allowing users to scale training up and out without infrastructure friction or permission sprawl. W&B Launch has supported access to high-performance compute required for building and deploying AI models on AWS, Google Cloud, and Microsoft Azure.
“W&B Launch gives our machine learning (ML) engineers easy access to compute so they can dramatically scale our training workloads for our computer vision models,” said Jayden Elliott, software engineer at VisualCortex, a video intelligence platform. “The intuitive platform enables our team to concentrate on our fundamental responsibilities of model training and evaluation, alleviating concerns regarding infrastructure management.”
W&B Launch now supports NVIDIA DGX systems, allowing Weights & Biases customers to easily scale up ML experimentation and hyperparameter tuning activities leveraging the power of NVIDIA DGX AI supercomputing.
Weights & Biases is also unveiling one of the first integrations with the NVIDIA NIM inference microservices, designed to bridge the gap between the complex world of AI development and the operational needs of enterprise environments. NIM is built for portability and control, allowing models to be deployed across a variety of infrastructures, whether from a local workstation, in the cloud, or in on-premises data centers. Users tracking model artifacts in Weights & Biases can leverage W&B Launch to deploy to NIM. The integration streamlines the model deployment process, accelerating models into production.
Additionally, Weights & Biases and CoreWeave, an NVIDIA Partner Network Elite Cloud Partner, have collaborated to integrate the Launch platform with the CoreWeave Cloud. The CoreWeave team has also created a W&B Launch Agent application, which is available on the CoreWeave Application Catalog. The app allows CoreWeave users to easily configure and install the W&B Launch Agent on their CoreWeave Cloud GPU cluster, further streamlining ML developers’ ability to set up model experimentation and evaluation jobs.
“Our mission is to build the best tools for machine learning practitioners around the world, and that also means collaborating with the best partners,” said Lukas Biewald, CEO at Weights & Biases. “The new integrations with NVIDIA and CoreWeave technologies will enhance our customers’ ability to easily train, tune, analyze, and deploy AI models to drive massive value for their organizations.”
“Developers are looking for more efficient ways to quickly, reliably scale LLMs and generative AI applications,” said Manuvir Das, vice president of Enterprise Computing at NVIDIA. “With expanded support for NVIDIA DGX and NVIDIA NIM, Weights & Biases customers can scale up initial experimentation on leading LLMs and deploy AI applications across their infrastructure.”
To learn more about its AI Developer Platform and the new integrations, visit Weights & Biases on-site at GTC from March 18 to 21 at booth 1505 and G115 or request a meeting with the team.
About Weights & Biases
Weights & Biases is the leading AI developer platform supporting end-to-end MLOps and LLMOps workflows. Used by over 30 foundation model builders and 1,000 companies to productionize machine learning at scale including teams at OpenAI, Toyota, and Microsoft. Weights & Biases is part of the new standard of best practices for machine learning.