Highlights

  • The focus is to provide a complete solution for the full Machine Learning (ML) lifecycle for ML practitioners.
  • Both the partners have been certified as NVIDIA AI Accelerated and NVIDIA DGX-Ready software partners.

The leading developer-first MLOps platform, Weights & Biases (W&B), announced partnership with Run: ai, a compute orchestration for Artificial Intelligence (AI) workload leader. The focus of this partnership is to provide a complete solution for the whole Machine Learning (ML) lifecycle for ML practitioners. It includes everything from MLOps and experiment tracking to resource management at scale.

The partnership will work on building integrations between both platforms to provide MLOps platform owners and ML developers a seamless experience for GPU and MLOps orchestration for deep learning and ML workloads.

Both companies will help streamline AI projects for AI researchers and IT and MLOps teams that manage AI infrastructure. ML practitioners can leverage the MLOps capabilities offered by Weights & Biases like collaboration reports, hyperparameter optimization, and experiment and artefact tracking. They can also gain access to NVIDIA accelerated computing resources orchestrated by the Atlas Platform of Run: ai, which is together an experience.

Developers will now be able to check NVIDIA GPU utilization within the W&B dashboard. Also, they will be able to improve utilization with Run: ai’s orchestration and scheduling capabilities.

The MLOps platform owners can optimize GPU resource consumption and scheduling for ML practitioners. They will be able to provide a single ML system of record for keeping track of accurate ML experiments history, model history, and dataset versioning.

The partnership between Run: ai and Weights & Biases is helpful for ML researchers and companies leveraging NVIDIA GPUs. Both the partners have been certified as NVIDIA AI Accelerated and NVIDIA DGX-Ready software partners. NVIDIA AI Enterprise software will be used to build the integrations between the partners and help organizations maximize their investment in NVIDIA-accelerated systems for AI and ML.

“Putting AI into production requires enterprises to manage a broad range of processes, including data governance, experiment tracking, workload management, and compute orchestration,” said Scott McClellan, Senior Director of Data Science and MLOps at NVIDIA. “Pairing Weights & Biases and Run: ai MLOps software with NVIDIA accelerated systems and NVIDIA AI Enterprise software enables enterprises to effectively deploy intelligent applications that solve real-world business challenges.”

Experts’ Take

“We are excited to partner with Run.ai to provide data scientists and ML practitioners the best tools for their ML development workflow,” said Seann Gardiner, VP of Business Development of Weights & Biases. “The combination of the developer tools provided by Weights and Biases, the dynamic allocation and orchestration of compute resources from Run.ai, and the optimized hardware and software provided by NVIDIA gives ML teams everything they need to accelerate the development and deployment of AI in the enterprise.”

“Run: ai and Weights & Biases together give data scientists, and the IT teams that support them, a complete solution for the full ML lifecycle, from building models to training and inference in production,” said Omri Geller CEO and co-founder of Run: ai. “Companies building their AI infrastructure and tooling from scratch can use NVIDIA GPUs, Run: ai, and W&B and have everything they need to manage AI initiatives at scale.”