Skip to content

Model Management ​

Models are the core of the asset layer. The real measure of whether a project has delivered is the number of model versions and deployment records, not the number of training runs.

Where: Model Deployment → Model Management in the left navigation.

Where Models Come From ​

SourceDescription
Completed trainingA training job that runs its full step budget registers a model version automatically
Checkpoint exportA checkpoint exported during training is registered as a model, also labeled with the training progress at export

End-to-End Lineage ​

Every model version records:

FieldDescription
Model name and versionVersions are generated from the date and a sequence number, for example v20261020-01
Training jobThe ID of the training job that produced the model
Dataset versionsThe one or more dataset versions used for training
Training strategyThe unified long-horizon strategy identifier
Hyperparameter snapshotThe parameters locked when training was submitted
StepThe training step the model corresponds to
ProgressFor checkpoint models only, the training progress at export
Model familyACT / EasyDP / Pi 0.5 / R0, and so on

Deployment records link back to model versions, so you can trace lineage in either direction:

text
Dataset version → Training job → Model version → Deployment record → Inference record → Bad case

How Models Are Counted ​

Models are counted by model version, not by training job. A single training job can produce several model versions: the final model plus any number of checkpoints.

Model Actions ​

ActionDescription
DetailsView lineage, snapshots, and all versions
Inference testSelect a device and start an inference test directly. See Deployment Overview.
DeployStart a cloud or edge deployment
EditChange the model name and notes; lineage fields can't be edited
DeleteBefore deleting, confirm that no running deployment uses this version

Multiple Model Families ​

The same data can be used to train models from different families, which makes it easier to move across robot embodiments and frameworks. The supported artifact formats and devices are determined by R&D validation; model and device combinations that haven't been validated can't be deployed.

NexCore is evolving rapidly; the latest version prevails.NexCore Docs · Embodied AI Skill Learning Platform by Lumos Robotics