AI application deployment cloud
How to use
DGX Cloud Lepton.
Turning Python model code into a Photon service, deploying a Photon or container as a GPU endpoint, and operating bounded AI jobs and inference workloads in a managed workspace.
DGX Cloud Lepton is a NVIDIA AI application and inference platform for Python Photons, containers, endpoints, jobs, and GPU resources. Turning Python model code into a Photon service, deploying a Photon or container as a GPU endpoint, and operating bounded AI jobs and inference workloads in a managed workspace. This guide covers the whole path in one place: official access, a first session that produces something reviewable, the checks that make output trustworthy, and the permissions worth limiting before you connect real work.
AI should make the work easier to inspect. If the workflow removes the source, the owner, or the review step, redesign the workflow.
Access & setup
Find, install, and sign in to DGX Cloud Lepton
Get into the official DGX Cloud Lepton experience with the right account and a setup you understand.
- 01
Start at https://build.nvidia.com/lepton and confirm the domain before entering account or payment information.
- 02
Availability: DGX Cloud Lepton is a browser-managed cloud platform with the Python leptonai package and lep command-line interface for local development and cloud deployment.
- 03
A DGX Cloud Lepton workspace, current leptonai package, scoped workspace credentials, a licensed model and container, secure secrets, representative tests, resource quotas, and deployment ownership.
- 04
Sign in with the account you intend to keep using, then review plan, data, notification, and permission settings.
- 05
Run one low-risk test task before connecting sensitive files, repositories, or workspace data.
- Use official download pages.
- Review permissions during setup.
- Keep installers and applications updated.
First session
Your first useful DGX Cloud Lepton session
Learn the interaction loop using a small task with a clear outcome.
- 01
Define the service contract, users, model and code licenses, data class, resources, latency, concurrency, scaling, access, secrets, observability, cost, and rollback targets.
- 02
Build the smallest Photon or pinned container locally, expose only typed handlers or the required port, and test inputs, outputs, errors, health, and dependency reproducibility.
- 03
Push an immutable version, create a private test endpoint with scoped secrets, bounded replicas and resources, and capture logs, events, utilization, latency, failures, and spend under load.
- 04
Canary a reviewed endpoint update rather than rerunning production in place, monitor it, preserve the previous version, and verify access revocation, teardown, storage deletion, and stopped billing.
- State the outcome before the background.
- Provide the real source material.
- Review the result before expanding the task.
Quality control
Check the quality of DGX Cloud Lepton output
Turn a promising generation into an asset you can legally and safely ship.
- 01
Restate the brief, the deliverable format, and where the asset will appear before judging any output.
- 02
Inspect at full resolution: generated media hides its errors at thumbnail size.
- 03
Check the details these systems reliably get wrong — hands, text, logos, reflections, symmetry, and edge artefacts.
- 04
Confirm the licence and commercial-use terms for your plan, and check the output does not reproduce a recognisable style, mark, or likeness you have no right to use.
- 05
Review for accessibility and delivery: alt text, contrast, colour profile, file size, and the exact export format.
- Inspect at full resolution before approving anything.
- Confirm commercial-use rights before an asset reaches a client.
- Keep the prompt and settings with the asset so it can be reproduced.
Privacy & permissions
Use DGX Cloud Lepton safely
Use the platform without surrendering rights, likenesses, or client material you do not control.
- 01
Confirm you hold the rights to every reference image, voice, or clip before uploading it.
- 02
Check whether your plan trains on submitted content, and opt out before uploading client work.
- 03
Never upload a real person's likeness or voice without their explicit, documented consent.
- 04
Review the output for marks, styles, or likenesses you have no licence to reproduce.
- 05
Label synthetic media where your client, platform, or jurisdiction requires disclosure.
- Older material may call endpoints deployments and the product Lepton AI; current documentation uses NVIDIA DGX Cloud Lepton. A Photon simplifies packaging but does not remove responsibility for unsafe handlers, arbitrary code, dependency and model licenses, public exposure, secrets, logs, autoscaling, or GPU cost. Pin versions, use private access, isolate secrets, cap resources, and update with rollback.
- Follow your organisation's approved-use policy.
- Never treat fluent output as authorization to act.
Core workflows
Step-by-step ways to use DGX Cloud Lepton for the work it does best.
Each workflow below has its own outcome, steps and review checkpoints.
Package the smallest typed service locally, test it as an API, and promote an immutable Photon only after dependency and behavior review.
↓ 02 WorkflowOperate and update a DGX Cloud Lepton endpointManage private access, secrets, replicas, resources, logs, events, and immutable revisions without disruptive reruns.
↓Workflow 01
Turn Python model code into a Lepton Photon service
Package the smallest typed service locally, test it as an API, and promote an immutable Photon only after dependency and behavior review.
Outcome: Create a reproducible service artifact with a narrow interface, known rights, and production-ready failure behavior.
- 01
Define the handler contract, users, input and output schema, model and code licenses, data class, safety, latency, resources, dependencies, errors, and acceptance tests.
- 02
Create a minimal Photon with typed handlers, validate and bound inputs, exclude arbitrary execution and embedded secrets, and pin model, package, and system dependencies.
- 03
Run locally with representative normal, malformed, adversarial, large, concurrent, and failure cases while checking outputs, logs, memory, startup, and health behavior.
- 04
Build and inspect the artifact, record its ID and source revision, scan dependencies, verify no credentials or private data are included, and push it to the correct workspace.
- 05
Deploy privately, smoke-test the generated API contract, compare with local results, and retain the artifact, test evidence, license record, and rollback version.
What good use looks like
- Expose the smallest typed interface.
- Never bake secrets into a Photon.
- Pin every production dependency.
Workflow 02
Operate and update a DGX Cloud Lepton endpoint
Manage private access, secrets, replicas, resources, logs, events, and immutable revisions without disruptive reruns.
Outcome: Keep GPU inference observable, cost-bounded, recoverable, and isolated across workspaces and environments.
- 01
Separate workspaces and credentials by environment, assign least privilege, and define access, data handling, retention, SLOs, replica and resource ceilings, budgets, and incident ownership.
- 02
Create an endpoint from a reviewed Photon ID or pinned container, attach scoped secrets, keep access private, expose one required port, and configure startup allowance and replica bounds.
- 03
Monitor status, events, logs, replicas, GPU utilization, queue, latency, errors, output parity, storage, and spend using redacted metadata and actionable alerts.
- 04
Test bad artifacts, slow startup, memory exhaustion, malformed requests, scale limits, provider failure, secret rotation, access revocation, rollback, stop, and removal.
- 05
Canary an endpoint update instead of rerunning production in place, retain the prior revision, audit access and cost, remove retired resources, and verify billing cessation.
What good use looks like
- Use immutable Photon or image versions.
- Update production with a canary.
- Verify endpoint removal stops billing.
Official references
Check the current product documentation.
Features, plan limits, availability, and data controls change. These official pages are the starting points used for this guide.
- DGX Cloud Lepton CLI start ↗
- Lepton endpoint reference ↗
- Lepton run reference ↗
- LeptonAI framework repository ↗