Skip to content
Sign inRequest access

Products

GPU CloudAMD MI325X and NVIDIA H200 by the hour or on reservation.
Dedicated ClustersPreviewPrivate, interconnected GPU clusters for sustained workloads.
SandboxesPreviewIsolated, short-lived environments for developers and AI agents.
Serverless GPUPreviewGPU containers that scale with requests, down to zero.
Serverless InferenceOpen models behind one OpenAI-compatible endpoint.
Dedicated EndpointsReserved serving capacity for one model, billed per GPU-hour.
PlaygroundSend a prompt to any model your workspace can reach.
Model DeploymentsPreviewImmutable deployments with instant promote and rollback.
Fine-tuningLoRA fine-tunes on the open models in our catalog.
Model TrainingPreviewFoundation-model training runs, managed end to end.
EvaluationsPreviewBenchmark suites and model comparisons you can rerun.
DatasetsManaged and synthetic training data, versioned and ready to use.
Data ManagementPreviewDatasets, connections and storage, with lineage across the platform.
PipelinesPreviewVersioned DAGs that automate data, training and deployment.
Data RoomsPreviewSecure document sharing with guests, NDAs and an audit trail.
NokuvaPreviewThe AI-native design editor.
TavocPreviewA Rust-native code platform: IDE and desktop apps.
FoltracComing soonMulti-cloud deployment with autonomous operations.
W0rkTreeComing soonVersion control for agent fleets.
Identity & AccessMembers, roles, policies, SSO and audit for every workspace.
Billing & CostUsage, invoices, credits and budgets in one place.
Logs & MonitoringRequest logs, latency and error rates across services.
Project CanvasPreviewSee and change a project's resources on one live canvas.

Pages

Home/
Platform/platform
All products/products
Pricing/pricing
GPU pricing/pricing#compute
Inference services/services/inference
Model training services/services/model-training
Protocol design/services/protocol-design
Research partnerships/services/research-partnerships
Protocols/protocols
Datasets/datasets
Documentationdocs.legionedge.ai
Changelog/changelog
News/news
Research/research
Blog/blog
Articles/articles
Events/events
Startup Program/programs/startup
Ambassador Program/programs/ambassador
Partner Program/programs/partner
Event Program/programs/events
Company/company
Careers/careers
Partners/partners
Contact/contact
Statusstatus.legionedge.ai
Terms of Service/legal/terms
Privacy Policy/legal/privacy

Serverless GPU

Preview

Ship a GPU container. Pay for the seconds it runs.

Deploy a container image with a GPU from the canvas and get a URL. Instances start on demand, scale between the limits you set and stop when traffic does, so you pay only for the seconds they run.

Request accessView pricing
Pricing
Per GPU-second
Layer
Compute
Deploy
Container images, Services, Jobs, Project canvas, Environment variables, Secrets

Why Serverless GPU

GPU capacity that follows your traffic.

A serverless GPU service is a container image with a GPU attached and a URL in front of it. Instances start when requests arrive, scale between the limits you set and stop when traffic does, so idle time costs nothing.

  • Scale to zero

    Set minimum and maximum instances and concurrency per service.

  • From the canvas

    Stage a service with your other resources and apply it in one changeset.

  • Per-second metering

    GPU and vCPU time billed by the second it runs.

How it works

GPU containers that scale with requests, down to zero.

  1. Stage a service

    Add a serverless GPU service to your project canvas with its container image, GPU type and region. GPU support in the region is checked before you submit.

  2. Set the limits

    Choose the minimum and maximum instances and the concurrency for each one. A minimum of zero lets the service stop between requests.

  3. Apply the changeset

    Deploy the service with your other resources in one changeset, and get a URL for it.

  4. Scale with requests

    Instances start on demand, scale between your limits and stop when traffic does, metered by the second they run.

In the console

Everything in one place.

Serverless GPU lives in the LegionEdge cloud console, next to every other service in the workspace, under one account and one identity.

  • Service settings

    Everything that decides how a service scales, in one place.

    • Container image, GPU type and region
    • Minimum and maximum instances
    • Concurrency per instance

  • On the canvas

    The service is a node beside the resources it uses.

    • Staged, reviewed and deployed with the rest of the project
    • Variables and secrets referenced by name
    • Review shows the cost change before you deploy

  • Metering

    Usage is recorded from the time instances actually run.

    • GPU seconds and vCPU seconds, metered separately
    • Spend for the service on its own usage page
    • Budgets apply like any other compute

Pricing

Per GPU-second

Serverless GPU is metered by the second: GPU time and vCPU time for as long as instances run, and nothing while the service is scaled to zero.

Full pricing
  • Keep a minimum of instances running for steady traffic, or let the service scale to zero.
  • A service scaled to zero starts an instance when the next request arrives.
  • Budgets and spend limits cover serverless usage too.

Related products

Pairs well with Serverless GPU

Get started

Start with a workload.

Connect to the platform directly, or work with our team on a specific technical challenge.

Self-service

LegionEdge Cloud

Create a workspace, explore the available models, and connect your application. The documentation takes you through your first request.

  • Turn on only the services you need
  • Pay per token or per GPU-hour
  • One account, one bill
Create a workspaceRead the docs

Custom engagements

Work with our team

Bring your data requirements, model goals, or infrastructure constraints. We'll define the scope, deliverables, and next steps together.

  • Scoped with an evaluation baseline
  • Dedicated clusters and reserved GPUs
  • A named engineer from day one
Discuss your projectsales@legionedge.ai