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

Pipelines

Preview

From new data to a new deployment, on its own.

Build pipelines as versioned graphs of steps and start them on a schedule, an event, a webhook, by hand or from another pipeline. Automate ingestion, transformation and validation, and close the loop from data to fine-tune, evaluation and deploy.

Request accessView pricing
Pricing
Billed as the services each step uses
Layer
Data
Starts on
A schedule, A platform event, A signed webhook, A manual run, Another pipeline

Why Pipelines

Close the loop from data to deploy.

A pipeline is a versioned graph of steps. Ingest and validate data, fine-tune on it, evaluate the result and promote it only when it clears the bar, with a person in the loop where you want one.

  • Visual editor

    Draft, validate and publish versions on a canvas.

  • Triggers

    Schedules, events, webhooks and manual runs.

  • Approval gates

    Hold a run until the right people approve it.

How it works

Versioned DAGs that automate data, training and deployment.

  1. Draw the graph

    Drop steps from the palette onto the canvas and connect typed ports. A connection between incompatible ports is refused, with the reason.

  2. Validate and publish

    Cycles, missing config and unconnected inputs are caught as you edit. Publishing freezes a version, and every run pins one.

  3. Trigger it

    Run on a cron schedule, when a dataset version is ready or a deployment fails, from a signed webhook, by hand or from another pipeline.

  4. Watch every run

    Follow the graph live as steps run, read each step’s logs and outputs, and retry only the steps that failed.

In the console

Everything in one place.

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

  • Step library

    Ready-made steps for each stage, each one a self-contained module with typed inputs and outputs.

    • Sources, SQL and Python transforms, splits and dedupe
    • Schema checks, expectations and PII scans
    • Fine-tune, evaluate, batch inference and deploy

  • Control flow

    Conditions, branches and fan-out, with gates where a decision belongs to a person.

    • Condition gates over earlier steps’ outputs
    • Approval gates with assignees and a timeout
    • Fan-out over a list, with a parallelism cap

  • Runs

    Every run, the trigger and version behind it, how long it took and what it cost.

    • Streaming logs for each step
    • Retry failed steps, cancel, or re-run with the same inputs
    • Notifications by email, Slack or webhook

  • Guardrails

    Pipelines run with exactly the access you give them, and no more.

    • Each run acts as the pipeline’s service account
    • Secrets are picked from secret groups, never typed in
    • Concurrency, timeouts and retries per pipeline

Pricing

Billed as the services each step uses

Pipelines are billed as the services each step uses: a fine-tune step as fine-tuning, a batch inference step as inference, a container step as compute.

Full pricing
  • Each run records what it cost, step by step.
  • Orchestration itself carries no separate charge.

Related products

Pairs well with Pipelines

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