Viam ML Training API
Submit and manage cloud-hosted ML training jobs that produce models deployable to viam-server. Supports built-in TFLite trainers and custom containerized trainers via SubmitTrainingJob and SubmitCustomTrainingJob.
OpenAPI Specification
openapi: 3.1.0
info:
title: Viam ML Training API
description: |
REST/JSON transcoding of the Viam MLTrainingService gRPC API. Submit, monitor, and
manage cloud-hosted ML training jobs that produce models deployable to viam-server.
Canonical contract: https://github.com/viamrobotics/api/blob/main/proto/viam/app/mltraining/v1/ml_training.proto
version: '2026.05'
contact:
name: Viam Support
url: https://www.viam.com/contact
servers:
- url: https://app.viam.com
description: Viam production cloud.
security:
- ApiKeyAuth: []
tags:
- name: Training Jobs
description: Built-in TFLite trainers and custom containerized trainers.
paths:
/viam.app.mltraining.v1.MLTrainingService/SubmitTrainingJob:
post:
summary: Viam Submit Training Job
description: Submit a built-in TFLite training job against a dataset.
operationId: submitTrainingJob
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [organization_id, dataset_id, model_name, model_type]
properties:
organization_id: { type: string }
dataset_id: { type: string }
model_name: { type: string }
model_version: { type: string }
model_type: { type: string, enum: [single_label_classification, multi_label_classification, object_detection] }
tags: { type: array, items: { type: string } }
responses:
'200':
description: Successful response.
content:
application/json:
schema:
type: object
properties:
id: { type: string }
/viam.app.mltraining.v1.MLTrainingService/SubmitCustomTrainingJob:
post:
summary: Viam Submit Custom Training Job
description: Submit a custom containerized training job (registry container) against a dataset.
operationId: submitCustomTrainingJob
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [organization_id, dataset_id, registry_item_id, model_name]
properties:
organization_id: { type: string }
dataset_id: { type: string }
registry_item_id: { type: string }
registry_item_version: { type: string }
model_name: { type: string }
model_version: { type: string }
arguments: { type: object }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/GetTrainingJob:
post:
summary: Viam Get Training Job
description: Retrieve a training job by id.
operationId: getTrainingJob
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [id]
properties:
id: { type: string }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/ListTrainingJobs:
post:
summary: Viam List Training Jobs
description: List training jobs scoped to an organization.
operationId: listTrainingJobs
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [organization_id]
properties:
organization_id: { type: string }
status: { type: string, enum: [unspecified, pending, in_progress, completed, failed, canceled, canceling] }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/CancelTrainingJob:
post:
summary: Viam Cancel Training Job
description: Cancel a pending or running training job.
operationId: cancelTrainingJob
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [id]
properties:
id: { type: string }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/DeleteCompletedTrainingJob:
post:
summary: Viam Delete Completed Training Job
description: Permanently delete a completed training job and its artifacts.
operationId: deleteCompletedTrainingJob
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [id]
properties:
id: { type: string }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/GetTrainingJobLogs:
post:
summary: Viam Get Training Job Logs
description: Retrieve logs for a training job.
operationId: getTrainingJobLogs
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema:
type: object
required: [id]
properties:
id: { type: string }
responses:
'200':
description: Successful response.
/viam.app.mltraining.v1.MLTrainingService/ListSupportedContainers:
post:
summary: Viam List Supported Containers
description: List the container images supported as custom training environments.
operationId: listSupportedContainers
tags: [Training Jobs]
requestBody:
required: true
content:
application/json:
schema: { type: object }
responses:
'200':
description: Successful response.
components:
securitySchemes:
ApiKeyAuth:
type: apiKey
in: header
name: key
description: Viam API key.