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

Model Training

Preview

Your expertise, in a model.

Train a model from scratch or continue pre-training on your data. We plan the run with you, operate the cluster, and hand over checkpoints you own.

Request accessView pricing
Pricing
From $2,500 / B params
Layer
Training
Starts from
Your data, LE-Code, LE-Interface, LE-Reasoning, Synthetic data, Open base models, Dedicated clusters

Why Model Training

A useful model is more than a checkpoint.

Know what it learned, how it was measured and how to run it. We plan the run with you, operate the cluster through it, and hand over weights and checkpoints that belong to you.

  • Planned with you

    Architecture, data mix and budget agreed up front, with a baseline and acceptance measures set before any compute is committed.

  • Operated for you

    Checkpointing, restarts and evaluation on our side, on a cluster our team operates through the run.

  • Yours to keep

    Weights, checkpoints and evaluation results belong to you. Training engagements are work for hire: we keep the methods, you keep the model.

How it works

From the task you define to the model you keep.

  1. Define the target

    Choose the task, base-model candidates, deployment constraints and acceptance measures, and establish a baseline before committing compute to training.

  2. Prepare the data

    Review permissions, quality and coverage. Curate the training mix and hold out evaluation data so the comparison stays meaningful.

  3. Train and compare

    We run the agreed experiments, inspect failures and compare candidates against the baseline before deciding which one to advance.

  4. Transfer the work

    Review the results together and take the agreed artifacts, evaluation tools and run notes, with the next decision made explicit.

What you receive

Everything you need to keep building.

Every engagement ends with artifacts your team can run, measure and extend, not just a set of weights.

  • A dataset built for the task

    Curated examples, documented preparation, and separate training and evaluation sets.

    • Your data, our corpora or synthetic data
    • Permissions, quality and coverage reviewed first
    • Held-out data for an honest comparison

  • Model artifacts and run history

    The agreed weights, configurations and training logs.

    • Weights and checkpoints that you own
    • Configurations and run logs
    • Base-model dependencies and licensing documented

  • An evaluation you can rerun

    A task-specific harness, baseline comparisons and an error analysis.

    • See where the model improved
    • See where it regressed
    • Know which behaviors still need work

  • A practical deployment handoff

    Inference examples, environment requirements, known limitations and recommended follow-up work.

    • Serve the checkpoint on a dedicated endpoint
    • Serving optimization scoped alongside, if you need it

Pricing

From $2,500 / B params

Managed training runs are priced per billion parameters, and we plan and operate the run. Distilling a larger model into a smaller one, or optimizing one you already run, is priced the same way.

Full pricing
  • Managed foundation-model training is part of the Enterprise plan.
  • Engagements carry a minimum contract, stated on the pricing page.
  • Fine-tuning and evaluations are billed separately, per GPU-hour.

Related products

Pairs well with Model Training

Model Training

Scope a training run.

Start with the behavior you want to change. A description of the workflow and representative examples are enough to begin.

Request accessTalk to us