What happens when the world’s most advanced ML platform meets the world’s fastest GPU cluster.

Paperspace Gradient is a perfect complement to NVIDIA DGX, including the brand new DGX A100, for machine learning workflows. Gradient enables shared DGX resources across many users with ease while maintaining CI/CD, experiment lifecycle management, monitoring, and multi-node training.

Contact SalesRead the docs

System Specifications

The NVIDIA DGX A100 system specifications include:

  • Eight NVIDIA A100  Tensor Core GPUs, delivering 5 petaflops of AI power, with 320GB in total GPU memory with 12.4TB  per second in bandwidth.
  • Six NVIDIA NVSwitch™ interconnect fabrics with third-generation NVIDIA NVLink® technology for 4.8TB  per second of bi-directional bandwidth.
  • Nine Mellanox ConnectX-6 HDR 200Gb per second network interfaces, offering a total of 3.6Tb  per second of bi-directional  bandwidth.
  • Mellanox In-Network Computing and network acceleration engines such as RDMA, GPUDirect® and Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)™ to enable the highest performance and scalability.
  • 15TB  Gen4 NVMe internal storage, which is 2x faster than Gen3 NVMe SSDs.
  • NVIDIA DGX software stack, which includes optimized software for AI and data science workloads, delivering maximized performance and enabling enterprises to achieve a faster return on their investment in AI infrastructure.

Developer-first MLOps platform with native DGX support.
Gradient lets you effortlessly scale from local cluster to the cloud.

Contact Sales

Resource management

Go from signup to training a model in 30 seconds. Leverage pre-configured templates and library of sample projects.

Powerful infrastructure

Powerful, low-cost GPUs you can launch with 1-click as well as infrastructure automation for on-prem clusters.

Scale instantly

Train in parallel and scale deployed models without any DevOps required. 1-click distributed training and hyperparemeters.

Full reproducibility

Automatic versioning, tagging, and life-cycle management. Build, evaluate, profile, and compare how different models perform.

Automated pipelines

Build powerful ML pipelines with modern, reproducible, and deterministic processes. Gradient is CI/CD for Machine Learning.

Collaboration and Insights

Improve visibility into team performance. Share projects with colleagues or leverage public projects.

And much more...

  • Model parsing and aggregation
  • System metrics
  • Team management
  • Pull request metrics
  • Custom container support
  • Dataset tracking
  • IDE
  • Unified logs
  • Private registries
  • Python CLI and SDK
  • Inference load-balancing
  • Tag management
  • Model repo
  • Native TensorBoard integration
  • gRPC & MPI support
Contact Sales

A modern ML stack built for the future

Full end-to-end ML for your entire team.
Gradient lets you move faster by ensuring that every step in your process is reproducible.

NEW! Gradient partners with NVIDIA
View the blog post announcing our new partnership with NVIDIA. Learn more