Last updated 07-30-2026
Category:
Reviews:
Join thousands of AI enthusiasts in the World of AI!
NVIDIA DGX Cloud Lepton (formerly Lepton AI)
NVIDIA DGX Cloud Lepton connects developers to GPU compute across a global network of cloud partners from one platform. You build, train, and deploy models with a single workflow whether the GPUs sit on AWS, CoreWeave, Lambda, or regional providers that meet data sovereignty rules. The site integrates NVIDIA NIM microservices, serverless endpoints on build.nvidia.com, and tools for inference, testing, and training without rearchitecting when you switch providers.
Independent GPU marketplaces make you pick one cloud and rewrite deployment scripts when capacity moves. DGX Cloud Lepton acts as a compute broker: discover GPUs from 25+ partners, run workloads where your data lives, and keep the same developer experience from prototype through production. NVIDIA acquired the original Lepton AI startup in April 2025 and folded it into this unified multi-cloud layer.
AI-native startups, model builders, and platform teams that outgrow a single cloud contract are the target users. Partners listed on the platform include AWS, CoreWeave, Crusoe, Lambda, Nebius, Scaleway, and Together AI with Blackwell and H100-class hardware. Access starts through build.nvidia.com APIs and the open-source leptonai Python library.
Unified development, training, and inference workflow across NVIDIA Cloud Partners and GPU marketplaces
Instant access to NVIDIA accelerated APIs and NIM microservices through build.nvidia.com
Deploy AI workloads across multi-cloud environments without rearchitecting when providers change
Run compute in specific regions to meet data sovereignty and low-latency requirements
Partner network includes AWS, CoreWeave, Lambda, Nebius, Scaleway, Together AI, and Mistral AI GPUs
Open-source leptonai Python library and lep CLI for managing endpoints, dev pods, and batch jobs
Integrated tools streamline the path from prototype to production on tens of thousands of GPUs
One workflow spans multiple GPU clouds without rewriting deployment code when capacity shifts.
Direct integration with NVIDIA NIM microservices and build.nvidia.com serverless endpoints.
Regional partner choice helps meet data residency requirements.
Open-source leptonai library remains maintained for local development and deployment.
No self-serve public pricing on lepton.ai; enterprise GPU access goes through NVIDIA and partners.
The independent Lepton AI product is gone after the 2025 NVIDIA acquisition.
Best fit for teams already in the NVIDIA ecosystem rather than casual hobbyists.
What is NVIDIA DGX Cloud Lepton?
NVIDIA DGX Cloud Lepton is a unified AI platform that connects developers to GPU compute from a global network of cloud providers. DGX Cloud Lepton offers one consistent experience for development, training, and inference across multi-cloud environments.
What happened to Lepton AI?
NVIDIA acquired Lepton AI in April 2025 and rebranded the platform as NVIDIA DGX Cloud Lepton. The lepton.ai domain now hosts the NVIDIA product while the open-source leptonai library remains available on GitHub.
Which cloud providers work with DGX Cloud Lepton?
DGX Cloud Lepton partners include AWS, CoreWeave, Crusoe, Lambda, Nebius, Nscale, Scaleway, Together AI, Mistral AI, and other NVIDIA Cloud Partners. The marketplace lists Blackwell, H100, and H200-class GPUs across regions.
How do developers access DGX Cloud Lepton?
Developers start through build.nvidia.com for serverless NVIDIA APIs and NIM microservices, then scale on DGX Cloud Lepton for multi-cloud GPU workloads. The leptonai Python package and lep CLI manage endpoints, dev pods, and batch jobs.
Can DGX Cloud Lepton run workloads in specific regions?
Yes. NVIDIA DGX Cloud Lepton lets teams bring compute from specific regions to meet data sovereignty rules and low-latency needs. You choose providers and regions on the Lepton platform without changing your application workflow.
Does DGX Cloud Lepton support inference and training?
Yes. DGX Cloud Lepton covers development, training, and inference in one platform with integrated services for testing, training, and deployment. NVIDIA NIM microservices handle inference while partner GPUs cover large-scale training.