Nebius Integrations

NVIDIA DGX Cloud Lepton is an advanced AI platform that facilitates connections for developers to a worldwide network of GPU computing resources across various cloud providers, all through a singular interface. It provides a cohesive experience for discovering and leveraging GPU capabilities, complemented by integrated AI services that enhance the deployment lifecycle across multiple cloud environments. With immediate access to NVIDIA's accelerated APIs, developers can begin their projects using serverless endpoints and prebuilt NVIDIA Blueprints, along with GPU-enabled computing. When scaling becomes necessary, DGX Cloud Lepton ensures smooth customization and deployment through its expansive global network of GPU cloud providers. Furthermore, it allows for effortless deployment across any GPU cloud, enabling AI applications to operate within multi-cloud and hybrid settings while minimizing operational complexities, and it leverages integrated services designed for inference, testing, and training workloads. This versatility ultimately empowers developers to focus on innovation without worrying about the underlying infrastructure.

NVIDIA DGX Cloud Serverless Inference provides a cutting-edge, serverless AI inference framework designed to expedite AI advancements through automatic scaling, efficient GPU resource management, multi-cloud adaptability, and effortless scalability. This solution enables users to reduce instances to zero during idle times, thereby optimizing resource use and lowering expenses. Importantly, there are no additional charges incurred for cold-boot startup durations, as the system is engineered to keep these times to a minimum. The service is driven by NVIDIA Cloud Functions (NVCF), which includes extensive observability capabilities, allowing users to integrate their choice of monitoring tools, such as Splunk, for detailed visibility into their AI operations. Furthermore, NVCF supports versatile deployment methods for NIM microservices, granting the ability to utilize custom containers, models, and Helm charts, thus catering to diverse deployment preferences and enhancing user flexibility. This combination of features positions NVIDIA DGX Cloud Serverless Inference as a powerful tool for organizations seeking to optimize their AI inference processes.