NVIDIA DGX Cloud

NVIDIA managed AI platform (DGX Cloud Lepton) for dev pods, training jobs and inference endpoints.

Provider: default
Kind: ML platform
Category: Compute
Official docs: https://docs.nvidia.com/dgx-cloud/lepton/get-started/index.html

Properties · 2 set by this component

property	type	values	required	kind default	this component
computeInstances	integer	—	no	—	inherited
computeSource · component property	enum	nvidia-cloud-partner \| bring-your-own-compute	no	nvidia-cloud-partner	inherited
experimentTracking	boolean	—	no	true	inherited
gpuEnabled	boolean	—	no	—	▸ true
managed	boolean	—	no	false	▸ true
modelRegistry	boolean	—	no	true	inherited
publicAccess	boolean	—	no	—	inherited
workload · component property	enum	dev-pods \| batch-jobs \| endpoints	no	—	inherited

Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

Stateful node without zone redundancyerror

stateful-node-without-zone-redundancy

A stateful compute node pinned to exactly one availability zone disappears with that zone, and its state goes with it. Nodes inside a cell are exempt, because there the cell is deliberately single-zone and redundancy comes from the other cells.

— all of these hold:
— the node's `stateless` is false
— the node sits inside exactly 1 availability-zone group
— this does not hold:
— the node sits inside at least one cell group

Use it in a design

nodes:
  - { id: "nvidia-dgx-cloud", component: "nvidia-dgx-cloud" }

Other ML platform products · 61

Platform for AIalibaba-paialibabacloudml-platform AWS Clean Rooms MLaws-clean-rooms-mlawsml-platform Amazon SageMaker AIaws-sagemakerawsml-platform Amazon SageMaker Canvasaws-sagemaker-canvasawsml-platform Amazon SageMaker Feature Storeaws-sagemaker-feature-storeawsml-platform Managed MLflow on Amazon SageMaker AIaws-sagemaker-mlflowawsml-platform Amazon SageMaker Unified Studioaws-sagemaker-unified-studioawsml-platform Microsoft Foundryazure-ai-foundryazureml-platform Azure Machine Learningazure-machine-learningazureml-platform Abacus.AIabacus-aiml-platform Alteryxalteryxml-platform Argillaargillaml-platform C3 AI Platformc3-aiml-platform ClearMLclearmlml-platform Cometcomet-mlml-platform CVATcvatml-platform Dataikudataikuml-platform DataRobotdatarobotml-platform Datasaurdatasaurml-platform Determined AIdeterminedml-platform Domino Data Labdomino-data-labml-platform DVCdvcml-platform Encordencordml-platform Feastfeastml-platform Flyteflyteml-platform H2O.aih2oml-platform Kilikiliml-platform KNIMEknimeml-platform Kubeflowkubeflowml-platform Label Studiolabel-studioml-platform Labelboxlabelboxml-platform Metaflowmetaflowml-platform Generic ML platformml-platformml-platform MLflowmlflowml-platform neptune.aineptune-aiml-platform Pachydermpachydermml-platform Polyaxonpolyaxonml-platform Prodigyprodigyml-platform RapidMiner AI Hubrapidminerml-platform Roboflowroboflowml-platform

and 21 more → ML platform page

Source

1# yaml-language-server: $schema=../../schema/component.schema.json
2# NVIDIA DGX Cloud: NVIDIA's managed AI platform, now DGX Cloud Lepton, for developing, training and deploying models
3# on GPUs from NVIDIA's cloud partners or your own clusters. Per "Welcome to NVIDIA DGX Cloud Lepton", it offers dev
4# pods (SSH, Jupyter, VS Code), batch jobs for distributed training, inference endpoints (NIM or your own container) and
5# node groups. Typed ml-platform, not virtual-machine, because the unit you draw is the workspace.
6schemaVersion: "1"
7type: "component"
8id: "nvidia-dgx-cloud"
9kind: "ml-platform"
10provider: "default"
11docs: "https://docs.nvidia.com/dgx-cloud/lepton/get-started/index.html"
12aliases: ["DGX Cloud"]
13defaults:
14  managed: true
15  gpuEnabled: true
16properties:
17  workload:
18    type: "enum"
19    values: ["dev-pods", "batch-jobs", "endpoints"]
20  computeSource:
21    type: "enum"
22    values: ["nvidia-cloud-partner", "bring-your-own-compute"]
23    default: "nvidia-cloud-partner"

The same page as markdown, for an agent: /catalogue/components/nvidia-dgx-cloud.md