API docs by Redocly

NVIDIA TAO API (6.25.10)

Download OpenAPI specification:

NVIDIA TAO (Train, Adapt, Optimize) API document

AUTHENTICATION

Endpoints related to User Authentication

Authenticate user with NGC credentials and set telemetry preferences

Authenticates a user using their NGC API key and organization name. Returns JWT token and user credentials upon successful authentication. The token can be used for subsequent API requests.

Authorizations:
api-key
Request Body schema: application/json
required

Login credentials including:

  • ngc_key: NGC API key for authentication
  • ngc_org_name: Organization name in NGC
  • enable_telemetry: Optional telemetry preference (default: False)
ngc_key
string <regex> <= 1000 characters
ngc_org_name
string <regex> <= 1000 characters
enable_telemetry
boolean or null

Responses

Request samples

Content type
application/json
{
  • "ngc_key": "/regex/",
  • "ngc_org_name": "/regex/",
  • "enable_telemetry": true
}

Response samples

Content type
application/json
{
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "token": "/regex/",
  • "user_name": "/regex/",
  • "user_email": "/regex/"
}

DATASET

Endpoints related to Datasets

Given dataset type return dataset formats or return all formats

Given dataset type return dataset formats or return all formats

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Responses

Response samples

Content type
application/json
{
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ]
}

List all accessible datasets

Returns a list of datasets that the authenticated user can access. Results can be filtered and paginated using query parameters. This includes:

  • Datasets owned by the user
  • Datasets shared with the user
  • Public datasets
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Organization name to list datasets from

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Number of records to skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Maximum number of records to return per page

sort
string
Enum: "date-descending" "date-ascending" "name-descending" "name-ascending"

Sort order for the results

name
string <= 5000 characters .*

Filter datasets by name (case-sensitive partial match)

format
string
Enum: "kitti" "pascal_voc" "raw" "coco_raw" "unet" "coco" "lprnet" "train" "test" "default" "custom" "classification_pyt" "visual_changenet_segment" "visual_changenet_classify"

Filter datasets by their format type

type
string
Enum: "object_detection" "segmentation" "image_classification" "character_recognition" "action_recognition" "pointpillars" "pose_classification" "ml_recog" "ocdnet" "ocrnet" "optical_inspection" "re_identification" "centerpose"

Filter datasets by their primary type

Responses

Response samples

Content type
application/json
{
  • "datasets": [
    ],
  • "pagination_info": {
    }
}

Delete multiple Datasets

Deletes multiple datasets and their associated resources. This endpoint:

  • Validates the datasets exist and user has access
  • Validates the datasets are not public
  • Validates the datasets are not read-only
  • Validates the datasets are not in use by any experiments
  • Validates no running jobs are using the datasets
  • Deletes the dataset files and metadata
  • Updates user permissions
  • Returns the status for each dataset
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
dataset_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "dataset_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Create new Dataset

Returns the new Dataset

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
required

Initial metadata for new Dataset (type and format required)

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
object
version
string <regex> <= 10 characters
logo
string <url> <= 2048 characters
type
string
Enum: "action_recognition" "segmentation" "ocdnet" "stylegan_xl" "optical_inspection" "bevfusion" "pointpillars" "re_identification" "sparse4d" "visual_changenet_classify" "depth_net_mono" "maxine_dataset" "visual_changenet_segment" "centerpose" "image_classification" "ml_recog" "user_custom" "depth_net_stereo" "object_detection" "pose_classification" "not_restricted" "ocrnet" "vlm"
format
string
Enum: "GenericDataset" "coco_raw" "maxine_gaze" "custom" "ThreeDVLM" "coco" "IssacStereo" "llava" "NvCLIP" "NYUDV2" "NYUDV2Relative" "BaseMetricMonoDataset" "raw" "ssl" "Crestereo" "BaseRelativeMonoDataset" "coco_panoptic" "unet" "FSD" "lprnet" "Eth3d" "classification_pyt" "odvg" "default" "Kitti" "IsaacRealDataset" "monai" "Middlebury" "ovpkl" "kitti"
workspace
string or null <uuid> <= 36 characters
url
string <url> <= 2048 characters
cloud_file_path
string or null <regex> <= 2048 characters
client_url
string or null <regex> <= 2048 characters
client_id
string or null <regex> <= 2048 characters
client_secret
string or null <regex> <= 2048 characters
filters
string or null <regex> <= 2048 characters
status
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
use_for
Array of strings or null <= 3 items
Enum: "training" "evaluation" "testing"
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
skip_validation
boolean or null
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "docker_env_vars": {
    },
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "client_url": "/regex/",
  • "client_id": "/regex/",
  • "client_secret": "/regex/",
  • "filters": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Retrieve details of a specific dataset

Returns detailed information about a specific dataset including:

  • Basic metadata (name, description, creation date)
  • Dataset format and type
  • Access permissions
  • Associated jobs and their status
  • Available actions
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Organization name owning the dataset

dataset_id
required
string <uuid> <= 36 characters

Unique identifier of the dataset to retrieve

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Delete a specific dataset

Deletes a dataset and its associated resources. The operation will:

  • Remove dataset files and metadata
  • Update user permissions

Deletion is only allowed if:

  • User has write permissions
  • Dataset is not public
  • Dataset is not read-only
  • Dataset is not in use by any experiments
  • No running jobs are using the dataset
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Organization name owning the dataset

dataset_id
required
string <uuid> <= 36 characters

Unique identifier of the dataset to delete

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Update Dataset

Returns the updated Dataset

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID of Dataset to update

Request Body schema: application/json
required

Updated metadata for Dataset

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
object
version
string <regex> <= 10 characters
logo
string <url> <= 2048 characters
type
string
Enum: "action_recognition" "segmentation" "ocdnet" "stylegan_xl" "optical_inspection" "bevfusion" "pointpillars" "re_identification" "sparse4d" "visual_changenet_classify" "depth_net_mono" "maxine_dataset" "visual_changenet_segment" "centerpose" "image_classification" "ml_recog" "user_custom" "depth_net_stereo" "object_detection" "pose_classification" "not_restricted" "ocrnet" "vlm"
format
string
Enum: "GenericDataset" "coco_raw" "maxine_gaze" "custom" "ThreeDVLM" "coco" "IssacStereo" "llava" "NvCLIP" "NYUDV2" "NYUDV2Relative" "BaseMetricMonoDataset" "raw" "ssl" "Crestereo" "BaseRelativeMonoDataset" "coco_panoptic" "unet" "FSD" "lprnet" "Eth3d" "classification_pyt" "odvg" "default" "Kitti" "IsaacRealDataset" "monai" "Middlebury" "ovpkl" "kitti"
workspace
string or null <uuid> <= 36 characters
url
string <url> <= 2048 characters
cloud_file_path
string or null <regex> <= 2048 characters
client_url
string or null <regex> <= 2048 characters
client_id
string or null <regex> <= 2048 characters
client_secret
string or null <regex> <= 2048 characters
filters
string or null <regex> <= 2048 characters
status
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
use_for
Array of strings or null <= 3 items
Enum: "training" "evaluation" "testing"
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
skip_validation
boolean or null
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "docker_env_vars": {
    },
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "client_url": "/regex/",
  • "client_id": "/regex/",
  • "client_secret": "/regex/",
  • "filters": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Partial update Dataset

Returns the updated Dataset

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID of Dataset to update

Request Body schema: application/json
required

Updated metadata for Dataset

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
object
version
string <regex> <= 10 characters
logo
string <url> <= 2048 characters
type
string
Enum: "action_recognition" "segmentation" "ocdnet" "stylegan_xl" "optical_inspection" "bevfusion" "pointpillars" "re_identification" "sparse4d" "visual_changenet_classify" "depth_net_mono" "maxine_dataset" "visual_changenet_segment" "centerpose" "image_classification" "ml_recog" "user_custom" "depth_net_stereo" "object_detection" "pose_classification" "not_restricted" "ocrnet" "vlm"
format
string
Enum: "GenericDataset" "coco_raw" "maxine_gaze" "custom" "ThreeDVLM" "coco" "IssacStereo" "llava" "NvCLIP" "NYUDV2" "NYUDV2Relative" "BaseMetricMonoDataset" "raw" "ssl" "Crestereo" "BaseRelativeMonoDataset" "coco_panoptic" "unet" "FSD" "lprnet" "Eth3d" "classification_pyt" "odvg" "default" "Kitti" "IsaacRealDataset" "monai" "Middlebury" "ovpkl" "kitti"
workspace
string or null <uuid> <= 36 characters
url
string <url> <= 2048 characters
cloud_file_path
string or null <regex> <= 2048 characters
client_url
string or null <regex> <= 2048 characters
client_id
string or null <regex> <= 2048 characters
client_secret
string or null <regex> <= 2048 characters
filters
string or null <regex> <= 2048 characters
status
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
use_for
Array of strings or null <= 3 items
Enum: "training" "evaluation" "testing"
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
skip_validation
boolean or null
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "docker_env_vars": {
    },
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "client_url": "/regex/",
  • "client_id": "/regex/",
  • "client_secret": "/regex/",
  • "filters": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "type": "action_recognition",
  • "format": "GenericDataset",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "url": "string",
  • "cloud_file_path": "/regex/",
  • "actions": [
    ],
  • "jobs": {
    },
  • "client_url": "/regex/",
  • "status": "starting",
  • "use_for": [
    ],
  • "base_experiment_pull_complete": "starting",
  • "base_experiment": [
    ],
  • "skip_validation": true,
  • "authorized_party_nca_id": "/regex/",
  • "validation_details": { }
}

Retrieve Specs schema

Returns the Specs schema for a given action

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

action
required
string
Enum: "dataset_convert" "convert" "kmeans" "augment" "train" "evaluate" "prune" "retrain" "export" "gen_trt_engine" "trtexec" "inference" "annotation" "analyze" "validate" "auto_label" "calibration_tensorfile"

Action name

Responses

Response samples

Content type
application/json
{ }

Run Dataset Jobs

Asynchronously starts a dataset action and returns corresponding Job ID. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the requested action is supported
  • Validates the provided specs match the action schema
  • Creates a new job with the provided parameters
  • Queues the job for execution
  • Returns the Job ID for tracking and retrieval
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

Request Body schema: application/json
parent_job_id
string or null <uuid> <= 36 characters
action
string
Enum: "auto3dseg" "quantize" "batchinfer" "evaluate" "generate" "retrain" "inference" "dataset_convert" "dataset_convert_gaze" "distill" "validate" "augment" "train" "export" "convert" "prune" "analyze" "validate_images" "annotation" "gen_trt_engine" "annotation_format_convert" "validate_annotations" "auto_label"
name
string or null <regex> <= 500 characters
description
string or null <regex> <= 1000 characters
specs
any
num_gpu
integer or null <int64> [ 0 .. 9223372036854776000 ]
platform_id
string or null <uuid> <= 36 characters

Responses

Request samples

Content type
application/json
{
  • "parent_job_id": "dd63d5e7-6727-402c-8127-3fd32aa9072a",
  • "action": "auto3dseg",
  • "name": "/regex/",
  • "description": "/regex/",
  • "specs": null,
  • "num_gpu": 9223372036854776000,
  • "platform_id": "f41a0584-7f4a-481e-9327-001efe2ebd2c"
}

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

List Jobs for Dataset

Returns the list of Jobs for a given dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Retrieves the list of jobs from storage
  • Applies pagination and filtering based on query parameters
  • Returns the filtered and paginated list of jobs
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <= 36 characters .*

ID for Dataset

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Optional skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Optional size for pagination

sort
string
Enum: "date-descending" "date-ascending"

Optional sort

Responses

Response samples

Content type
application/json
{
  • "jobs": [
    ],
  • "pagination_info": {
    }
}

Delete multiple Dataset Jobs

Deletes multiple jobs within a dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the jobs exist and are deletable
  • Deletes the job files and metadata
  • Updates job status to 'deleted'
  • Returns the status for each job
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

Request Body schema: application/json
job_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "job_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Retry Dataset Jobs

Asynchronously retries a dataset action and returns corresponding Job ID. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists and is retryable
  • Creates a new job with the same parameters as the original job
  • Queues the job for execution
  • Returns the new Job ID for tracking and retrieval
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

Retrieve Job for Dataset

Returns the Job for a given dataset and job ID. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists
  • Retrieves the job from storage
  • Returns the job
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID of Dataset

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "parent_id": "1c6ca187-e61f-4301-8dcb-0e9749e89eef",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "action": "auto3dseg",
  • "status": "Done",
  • "job_details": {
    },
  • "specs": null,
  • "name": "/regex/",
  • "description": "/regex/",
  • "num_gpu": 9223372036854776000,
  • "platform_id": "f41a0584-7f4a-481e-9327-001efe2ebd2c",
  • "dataset_id": "8c4c51f1-f6f3-43bc-b65d-7415e8ef22c0"
}

Delete Dataset Job

Deletes a specific job within a dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists and is deletable
  • Deletes the job files and metadata
  • Updates job status to 'deleted'
  • Returns the deletion status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Retrieve Schema for a job

Returns the Specs schema for a given job. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists
  • Retrieves the schema for the job's action
  • Returns the schema
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

job_id
required
string <uuid> <= 36 characters

ID for JOB

Responses

Response samples

Content type
application/json
{ }

Get Job logs for Dataset

Returns the job logs for a given dataset and job ID. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists
  • Retrieves the job logs from storage
  • Returns the job logs
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID of Dataset

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
text/plain
Execution status: PASS

Cancel Dataset Job

Cancels a specific job within a dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists and is cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Cancel all Jobs under dataset

Cancels all jobs within a dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the jobs exist and are cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
  • Returns the cancellation status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID for Dataset

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Cancel all Jobs under multiple datasets

Cancels all jobs within multiple datasets. This endpoint:

  • Validates the datasets exist and user has access
  • Validates the jobs exist and are cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
  • Returns the cancellation status for each dataset
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
dataset_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "dataset_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Download Job Artifacts

Downloads all artifacts produced by a given job within a dataset. This endpoint:

  • Validates the dataset exists and user has access
  • Validates the job exists
  • Downloads all job files
  • Returns the downloaded files as a tarball
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

dataset_id
required
string <uuid> <= 36 characters

ID of Dataset

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

EXPERIMENT

Endpoints related to Experiments

List Experiments

Returns the list of Experiments

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Optional skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Optional size for pagination

sort
string
Enum: "date-descending" "date-ascending" "name-descending" "name-ascending"

Optional sort

name
string <= 5000 characters .*

Optional name filter

type
string
Enum: "vision" "medical"

Optional type filter

network_arch
string
Enum: "action_recognition" "classification_pyt" "mal" "ml_recog" "ocdnet" "ocrnet" "optical_inspection" "pointpillars" "pose_classification" "re_identification" "deformable_detr" "dino" "segformer" "visual_changenet_classify" "visual_changenet_segment" "centerpose"

Optional network architecture filter

read_only
boolean

Optional read_only filter

user_only
boolean

Optional filter to select user owned experiments only

tag
string <= 5000 characters .*

Optional tag filter

Responses

Response samples

Content type
application/json
{
  • "experiments": [
    ],
  • "pagination_info": {
    }
}

Delete multiple Experiments

Cancels all related running jobs and returns the status of deleted Experiments

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
experiment_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "experiment_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

Create new Experiment

Returns the new Experiment

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
required

Initial metadata for new Experiment (base_experiment or network_arch required)

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
model_description
string <regex> <= 1000 characters
version
string <regex> <= 500 characters
logo
string or null <url> <= 2048 characters
ngc_path
string <regex> <= 250 characters
workspace
string or null <uuid> <= 36 characters
object or null
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
additional_id_info
string or null <regex> <= 100 characters
object
checkpoint_choose_method
string
Enum: "latest_model" "best_model" "from_epoch_number"
object
encryption_key
string <regex> <= 100 characters
network_arch
string
Enum: "deformable_detr" "action_recognition" "monai_automl_generated" "grounding_dino" "ocdnet" "monai_vista2d" "stylegan_xl" "optical_inspection" "monai_genai" "monai_automl" "bevfusion" "pointpillars" "mal" "monai_custom" "re_identification" "vila" "sparse4d" "image" "monai_detection" "visual_changenet_classify" "monai_classification" "depth_net_mono" "maxine_eye_contact" "monai_vista3d" "monai_segmentation" "mae" "visual_changenet_segment" "mask2former" "monai_annotation" "classification_pyt" "segformer" "centerpose" "rtdetr" "nvdinov2" "ml_recog" "mask_grounding_dino" "monai_maisi" "analytics" "augmentation" "annotations" "dino" "depth_net_stereo" "pose_classification" "ocrnet" "auto_label" "cosmos-rl"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
eval_dataset
string <uuid> <= 36 characters
inference_dataset
string <uuid> <= 36 characters
calibration_dataset
string <uuid> <= 36 characters
train_datasets
Array of strings <uuid> <= 9223372036854776000 items [ items <uuid > <= 36 characters ]
read_only
boolean
public
boolean
(object or null) or AutoML (object)
metric
string or null <regex> <= 100 characters
type
string
Enum: "vision" "medical" "maxine"
realtime_infer
boolean
object or null
bundle_url
string or null <regex> <= 1000 characters
realtime_infer_request_timeout
integer or null <int64> [ 0 .. 9223372036854776000 ]
Array of (object or null) or ExperimentActions (object) <= 9223372036854776000 items
tensorboard_enabled
boolean or null
tags
Array of strings <regex> <= 16 items [ items <regex > <= 36 characters ]
retry_experiment_id
string or null <uuid> <= 36 characters
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "docker_env_vars": {
    },
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "automl_settings": { },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "model_params": {
    },
  • "bundle_url": "/regex/",
  • "realtime_infer_request_timeout": 9223372036854776000,
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "retry_experiment_id": "9156f81c-d614-4eaa-af09-2507d8dc3481",
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

List Experiments that can be used for transfer learning

Returns the list of models published in NGC public catalog and private org's model registry

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Optional skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Optional size for pagination

sort
string
Enum: "date-descending" "date-ascending" "name-descending" "name-ascending"

Optional sort

name
string <= 5000 characters .*

Optional name filter

type
string
Enum: "vision" "medical"

Optional type filter

network_arch
string
Enum: "action_recognition" "classification_pyt" "mal" "ml_recog" "ocdnet" "ocrnet" "optical_inspection" "pointpillars" "pose_classification" "re_identification" "deformable_detr" "dino" "segformer" "visual_changenet_classify" "visual_changenet_segment" "centerpose"

Optional network architecture filter

read_only
boolean

Optional read_only filter

Responses

Response samples

Content type
application/json
{
  • "experiments": [
    ],
  • "pagination_info": {
    }
}

Retrieve Experiment

Returns the Experiment

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment to return

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

Delete Experiment

Cancels all related running jobs and returns the deleted Experiment

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment to delete

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

Update Experiment

Updates an existing experiment with new metadata. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the provided metadata matches the schema
  • Updates the experiment metadata in storage
  • Returns the updated experiment metadata
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment to update

Request Body schema: application/json
required

Updated metadata for Experiment

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
model_description
string <regex> <= 1000 characters
version
string <regex> <= 500 characters
logo
string or null <url> <= 2048 characters
ngc_path
string <regex> <= 250 characters
workspace
string or null <uuid> <= 36 characters
object or null
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
additional_id_info
string or null <regex> <= 100 characters
object
checkpoint_choose_method
string
Enum: "latest_model" "best_model" "from_epoch_number"
object
encryption_key
string <regex> <= 100 characters
network_arch
string
Enum: "deformable_detr" "action_recognition" "monai_automl_generated" "grounding_dino" "ocdnet" "monai_vista2d" "stylegan_xl" "optical_inspection" "monai_genai" "monai_automl" "bevfusion" "pointpillars" "mal" "monai_custom" "re_identification" "vila" "sparse4d" "image" "monai_detection" "visual_changenet_classify" "monai_classification" "depth_net_mono" "maxine_eye_contact" "monai_vista3d" "monai_segmentation" "mae" "visual_changenet_segment" "mask2former" "monai_annotation" "classification_pyt" "segformer" "centerpose" "rtdetr" "nvdinov2" "ml_recog" "mask_grounding_dino" "monai_maisi" "analytics" "augmentation" "annotations" "dino" "depth_net_stereo" "pose_classification" "ocrnet" "auto_label" "cosmos-rl"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
eval_dataset
string <uuid> <= 36 characters
inference_dataset
string <uuid> <= 36 characters
calibration_dataset
string <uuid> <= 36 characters
train_datasets
Array of strings <uuid> <= 9223372036854776000 items [ items <uuid > <= 36 characters ]
read_only
boolean
public
boolean
(object or null) or AutoML (object)
metric
string or null <regex> <= 100 characters
type
string
Enum: "vision" "medical" "maxine"
realtime_infer
boolean
object or null
bundle_url
string or null <regex> <= 1000 characters
realtime_infer_request_timeout
integer or null <int64> [ 0 .. 9223372036854776000 ]
Array of (object or null) or ExperimentActions (object) <= 9223372036854776000 items
tensorboard_enabled
boolean or null
tags
Array of strings <regex> <= 16 items [ items <regex > <= 36 characters ]
retry_experiment_id
string or null <uuid> <= 36 characters
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "docker_env_vars": {
    },
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "automl_settings": { },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "model_params": {
    },
  • "bundle_url": "/regex/",
  • "realtime_infer_request_timeout": 9223372036854776000,
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "retry_experiment_id": "9156f81c-d614-4eaa-af09-2507d8dc3481",
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

Partial update Experiment

Partially updates an existing experiment with new metadata. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the provided metadata matches the schema
  • Updates the experiment metadata in storage
  • Returns the updated experiment metadata
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment to update

Request Body schema: application/json
required

Updated metadata for Experiment

name
string <regex> <= 500 characters
shared
boolean
user_id
string <uuid> <= 36 characters
description
string <regex> <= 1000 characters
model_description
string <regex> <= 1000 characters
version
string <regex> <= 500 characters
logo
string or null <url> <= 2048 characters
ngc_path
string <regex> <= 250 characters
workspace
string or null <uuid> <= 36 characters
object or null
base_experiment_pull_complete
string
Enum: "starting" "in_progress" "pull_complete" "invalid_pull"
additional_id_info
string or null <regex> <= 100 characters
object
checkpoint_choose_method
string
Enum: "latest_model" "best_model" "from_epoch_number"
object
encryption_key
string <regex> <= 100 characters
network_arch
string
Enum: "deformable_detr" "action_recognition" "monai_automl_generated" "grounding_dino" "ocdnet" "monai_vista2d" "stylegan_xl" "optical_inspection" "monai_genai" "monai_automl" "bevfusion" "pointpillars" "mal" "monai_custom" "re_identification" "vila" "sparse4d" "image" "monai_detection" "visual_changenet_classify" "monai_classification" "depth_net_mono" "maxine_eye_contact" "monai_vista3d" "monai_segmentation" "mae" "visual_changenet_segment" "mask2former" "monai_annotation" "classification_pyt" "segformer" "centerpose" "rtdetr" "nvdinov2" "ml_recog" "mask_grounding_dino" "monai_maisi" "analytics" "augmentation" "annotations" "dino" "depth_net_stereo" "pose_classification" "ocrnet" "auto_label" "cosmos-rl"
base_experiment
Array of strings <uuid> <= 2 items [ items <uuid > <= 36 characters ]
eval_dataset
string <uuid> <= 36 characters
inference_dataset
string <uuid> <= 36 characters
calibration_dataset
string <uuid> <= 36 characters
train_datasets
Array of strings <uuid> <= 9223372036854776000 items [ items <uuid > <= 36 characters ]
read_only
boolean
public
boolean
(object or null) or AutoML (object)
metric
string or null <regex> <= 100 characters
type
string
Enum: "vision" "medical" "maxine"
realtime_infer
boolean
object or null
bundle_url
string or null <regex> <= 1000 characters
realtime_infer_request_timeout
integer or null <int64> [ 0 .. 9223372036854776000 ]
Array of (object or null) or ExperimentActions (object) <= 9223372036854776000 items
tensorboard_enabled
boolean or null
tags
Array of strings <regex> <= 16 items [ items <regex > <= 36 characters ]
retry_experiment_id
string or null <uuid> <= 36 characters
authorized_party_nca_id
string or null <regex> <= 2048 characters

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "docker_env_vars": {
    },
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "automl_settings": { },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "model_params": {
    },
  • "bundle_url": "/regex/",
  • "realtime_infer_request_timeout": 9223372036854776000,
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "retry_experiment_id": "9156f81c-d614-4eaa-af09-2507d8dc3481",
  • "authorized_party_nca_id": "/regex/"
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "description": "/regex/",
  • "model_description": "/regex/",
  • "version": "/regex/",
  • "logo": "string",
  • "ngc_path": "/regex/",
  • "workspace": "3f216741-15dd-4e46-b5ac-0077a2640e89",
  • "sha256_digest": {
    },
  • "base_experiment_pull_complete": "starting",
  • "additional_id_info": "/regex/",
  • "checkpoint_choose_method": "latest_model",
  • "checkpoint_epoch_number": {
    },
  • "encryption_key": "/regex/",
  • "network_arch": "deformable_detr",
  • "base_experiment": [
    ],
  • "dataset_type": "action_recognition",
  • "dataset_formats": [
    ],
  • "accepted_dataset_intents": [
    ],
  • "eval_dataset": "f06341d1-b6f7-4d94-99b1-6d57b5a471e6",
  • "inference_dataset": "287bd7ed-1072-49bc-8d0c-261f29925b9f",
  • "calibration_dataset": "06da9759-37ba-45ac-b600-dfa833bc04e5",
  • "train_datasets": [
    ],
  • "read_only": true,
  • "public": true,
  • "actions": [
    ],
  • "jobs": {
    },
  • "status": "Done",
  • "all_jobs_cancel_status": "Done",
  • "automl_settings": {
    },
  • "metric": "/regex/",
  • "type": "vision",
  • "realtime_infer": true,
  • "realtime_infer_support": true,
  • "model_params": {
    },
  • "realtime_infer_request_timeout": 86400,
  • "bundle_url": "/regex/",
  • "base_experiment_metadata": { },
  • "source_type": "ngc",
  • "experiment_actions": [
    ],
  • "tensorboard_enabled": true,
  • "tags": [
    ],
  • "authorized_party_nca_id": "/regex/"
}

Retrieve Specs schema

Returns the Specs schema for a given action and experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the action is supported
  • Retrieves the schema for the action
  • Returns the schema
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

action
required
string
Enum: "dataset_convert" "convert" "kmeans" "augment" "train" "evaluate" "prune" "retrain" "export" "gen_trt_engine" "trtexec" "inference" "annotation" "analyze" "validate" "auto_label" "calibration_tensorfile"

Action name

Responses

Response samples

Content type
application/json
{ }

Retrieve Base Experiment Specs schema

Returns the Specs schema for a given action of the base experiment. This endpoint:

  • Validates the base experiment exists and user has access
  • Validates the action is supported
  • Retrieves the schema for the action
  • Returns the schema
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Base Experiment

action
required
string
Enum: "train" "evaluate" "prune" "retrain" "export" "gen_trt_engine" "trtexec" "inference" "auto_label"

Action name

Responses

Response samples

Content type
application/json
{ }

Run Experiment Jobs

Asynchronously starts a Experiment Action and returns corresponding Job ID. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the requested action is supported
  • Validates the provided specs match the action schema
  • Creates a new job with the provided parameters
  • Queues the job for execution
  • Returns the Job ID for tracking and retrieval
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

Request Body schema: application/json
parent_job_id
string or null <uuid> <= 36 characters
action
string
Enum: "auto3dseg" "quantize" "batchinfer" "evaluate" "generate" "retrain" "inference" "dataset_convert" "dataset_convert_gaze" "distill" "validate" "augment" "train" "export" "convert" "prune" "analyze" "validate_images" "annotation" "gen_trt_engine" "annotation_format_convert" "validate_annotations" "auto_label"
name
string or null <regex> <= 500 characters
description
string or null <regex> <= 1000 characters
specs
any
num_gpu
integer or null <int64> [ 0 .. 9223372036854776000 ]
platform_id
string or null <uuid> <= 36 characters

Responses

Request samples

Content type
application/json
{
  • "parent_job_id": "dd63d5e7-6727-402c-8127-3fd32aa9072a",
  • "action": "auto3dseg",
  • "name": "/regex/",
  • "description": "/regex/",
  • "specs": null,
  • "num_gpu": 9223372036854776000,
  • "platform_id": "f41a0584-7f4a-481e-9327-001efe2ebd2c"
}

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

List Jobs for Experiment

Returns the list of Jobs for a given experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Retrieves the list of jobs from storage
  • Applies pagination and filtering based on query parameters
  • Returns the filtered and paginated list of jobs
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <= 36 characters .*

ID for Experiment

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Optional skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Optional size for pagination

sort
string
Enum: "date-descending" "date-ascending"

Optional sort

Responses

Response samples

Content type
application/json
{
  • "jobs": [
    ],
  • "pagination_info": {
    }
}

Delete multiple Experiment Jobs

Deletes multiple jobs within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the jobs exist and are deletable
  • Deletes the job files and metadata
  • Updates job status to 'deleted'
  • Returns the deletion status for each job
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

Request Body schema: application/json
job_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "job_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Retry Experiment Jobs

Asynchronously retries a Experiment Action and returns corresponding Job ID. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is retryable
  • Creates a new job with the same parameters as the original job
  • Queues the job for execution
  • Returns the new Job ID for tracking and retrieval
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

Get epoch numbers present for this job

Retrieves the epoch numbers for the checkpoints present for this job. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Retrieves the list of epoch numbers from storage
  • Returns the list of epoch numbers
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Organization name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Publish models to NGC

Publishes models to NGC private registry. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is publishable
  • Validates the provided metadata matches the schema
  • Publishes the model to NGC with the provided metadata
  • Returns a success message
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Request Body schema: application/json
display_name
string or null <regex> <= 500 characters
description
string or null <regex> <= 500 characters
team_name
string or null <regex> <= 500 characters

Responses

Request samples

Content type
application/json
{
  • "display_name": "/regex/",
  • "description": "/regex/",
  • "team_name": "/regex/"
}

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

Remove publish models from NGC

Removes models from NGC private registry. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is publishable
  • Validates the provided metadata matches the schema
  • Removes the model from NGC
  • Returns a success message
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Request Body schema: application/json
display_name
string or null <regex> <= 500 characters
description
string or null <regex> <= 500 characters
team_name
string or null <regex> <= 500 characters

Responses

Request samples

Content type
application/json
{
  • "display_name": "/regex/",
  • "description": "/regex/",
  • "team_name": "/regex/"
}

Response samples

Content type
application/json
"497f6eca-6276-4993-bfeb-53cbbbba6f08"

Retrieve Schema for a job

Returns the Specs schema for a given job. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Retrieves the schema for the job's action
  • Returns the schema
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for JOB

Responses

Response samples

Content type
application/json
{ }

Retrieve Job for Experiment

Returns the Job for a given experiment and job ID. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Retrieves the job from storage
  • Returns the job
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "parent_id": "1c6ca187-e61f-4301-8dcb-0e9749e89eef",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "action": "auto3dseg",
  • "status": "Done",
  • "job_details": {
    },
  • "sync": true,
  • "specs": null,
  • "name": "/regex/",
  • "description": "/regex/",
  • "num_gpu": 9223372036854776000,
  • "platform_id": "f41a0584-7f4a-481e-9327-001efe2ebd2c",
  • "experiment_id": "916afd89-cac5-4339-9c59-dd068abdfa69"
}

Delete Experiment Job

Deletes a specific job within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is deletable
  • Deletes the job files and metadata
  • Updates job status to 'deleted'
  • Returns the deletion status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Get Job logs for Experiment

Returns the job logs for a given experiment and job ID. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Retrieves the job logs from storage
  • Returns the job logs
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment

job_id
required
string <uuid> <= 36 characters

Job ID

query Parameters
automl_experiment_index
integer <int32> [ 0 .. 2147483647 ]

Optional filter to retrieve logs from specific autoML experiment

Responses

Response samples

Content type
text/plain
Execution status: PASS

Retrieve usable AutoML details

Retrieves usable AutoML details for a given experiment and job ID. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Retrieves the AutoML details from storage
  • Returns the AutoML details
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Pause Experiment Job - only for training

Pauses a specific job within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is pausable
  • Updates the job status to 'paused'
  • Persists status changes to storage
  • Triggers any necessary pause workflows
  • Returns the pause status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Cancel all Jobs under experiment

Cancels all jobs within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the jobs exist and are cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
  • Returns the cancellation status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Cancel all Jobs under multiple experiments

Cancels all jobs within multiple experiments. This endpoint:

  • Validates the experiments exist and user has access
  • Validates the jobs exist and are cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
  • Returns the cancellation status for each experiment
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
experiment_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "experiment_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Cancel Experiment Job or pause training

Cancels a specific job within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is cancellable
  • Updates the job status to 'cancelled'
  • Persists status changes to storage
  • Triggers any necessary cancellation workflows
  • Returns the cancellation status
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Resume Experiment Job

Resumes a specific job within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists and is resumable
  • Validates the provided metadata matches the schema
  • Updates the job metadata
  • Queues the job for execution
  • Returns the new Job ID for tracking and retrieval
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID for Experiment

job_id
required
string <uuid> <= 36 characters

ID for Job

Request Body schema: application/json
optional

Adjustable metadata for the resumed job.

parent_job_id
string or null <uuid> <= 36 characters
specs
any or null

Responses

Request samples

Content type
application/json
{
  • "parent_job_id": "dd63d5e7-6727-402c-8127-3fd32aa9072a",
  • "specs": null
}

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

Download Job Artifacts

Downloads the artifacts produced by a given job within an experiment. This endpoint:

  • Validates the experiment exists and user has access
  • Validates the job exists
  • Validates the requested export type is supported
  • Downloads the job artifacts
  • Returns the downloaded artifacts as a tarball
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

ID of Experiment

job_id
required
string <uuid> <= 36 characters

Job ID

Responses

Response samples

Content type
application/json
{
  • "error_desc": "/regex/",
  • "error_code": -9223372036854776000
}

ORGS

Retrieve available GPU types based on the backend during deployment

Retrieve available GPU types based on the backend during deployment

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Responses

Response samples

Content type
application/json
{ }

Retrieve available GPU types based on the backend during deployment

Retrieve available GPU types based on the backend during deployment

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Responses

Response samples

Content type
application/json
{ }

WORKSPACE

List workspaces

Returns the list of workspaces

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

query Parameters
skip
integer <int32> [ 0 .. 2147483647 ]

Optional skip for pagination

size
integer <int32> [ 0 .. 2147483647 ]

Optional size for pagination

sort
string
Enum: "date-descending" "date-ascending" "name-descending" "name-ascending"

Optional sort

name
string <= 5000 characters .*

Optional name filter

format
string
Enum: "monai" "unet" "custom"

Optional format filter

type
string
Enum: "object_detection" "segmentation" "image_classification"

Optional type filter

Responses

Response samples

Content type
application/json
[
  • {
    }
]

Delete multiple Workspaces

Cancels all related running jobs and returns the status of deleted Workspaces

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
workspace_ids
Array of strings <uuid> <= 2147483647 items [ items <uuid > <= 36 characters ]

Responses

Request samples

Content type
application/json
{
  • "workspace_ids": [
    ]
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Create new Workspace

Returns the new Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
required

Initial metadata for new Workspace (type and format required)

name
string <regex> <= 500 characters
shared
boolean
version
string <regex> <= 10 characters
cloud_type
string
Enum: "aws" "azure" "seaweedfs" "huggingface" "self_hosted"
cloud_specific_details
any

Responses

Request samples

Content type
application/json
Example
{
  • "cloud_details": {
    }
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Retrieve Workspace

Returns the Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

workspace_id
required
string <uuid> <= 36 characters

ID of Workspace to return

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Delete Workspace

Cancels all related running jobs and returns the deleted Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

workspace_id
required
string <uuid> <= 36 characters

ID of Workspace to delete

Responses

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Update Workspace

Returns the updated Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

workspace_id
required
string <uuid> <= 36 characters

ID of Workspace to update

Request Body schema: application/json
required

Updated metadata for Workspace

name
string <regex> <= 500 characters
shared
boolean
version
string <regex> <= 10 characters
cloud_type
string
Enum: "aws" "azure" "seaweedfs" "huggingface" "self_hosted"
cloud_specific_details
any

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws",
  • "cloud_specific_details": null
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Partial update Workspace

Returns the updated Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

workspace_id
required
string <uuid> <= 36 characters

ID of Workspace to update

Request Body schema: application/json
required

Updated metadata for Workspace

name
string <regex> <= 500 characters
shared
boolean
version
string <regex> <= 10 characters
cloud_type
string
Enum: "aws" "azure" "seaweedfs" "huggingface" "self_hosted"
cloud_specific_details
any

Responses

Request samples

Content type
application/json
{
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws",
  • "cloud_specific_details": null
}

Response samples

Content type
application/json
{
  • "id": "497f6eca-6276-4993-bfeb-53cbbbba6f08",
  • "user_id": "a169451c-8525-4352-b8ca-070dd449a1a5",
  • "created_on": "2019-08-24T14:15:22Z",
  • "last_modified": "2019-08-24T14:15:22Z",
  • "name": "/regex/",
  • "shared": true,
  • "version": "/regex/",
  • "cloud_type": "aws"
}

Retrieve datasets from Workspace

Returns the datasets matched with the request body args inside the Workspace

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

workspace_id
required
string <uuid> <= 36 characters

ID of Workspace to return

Responses

Response samples

Content type
application/json
{
  • "dataset_paths": [
    ]
}

WORKSPACES

Backup MongoDB data using workspace metadata

Backs up all MongoDB databases using provided workspace cloud credentials

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
workspace_metadata
required
object

Workspace metadata containing cloud credentials

backup_file_name
string

Optional backup file name

Responses

Request samples

Content type
application/json
{
  • "workspace_metadata": { },
  • "backup_file_name": "string"
}

Response samples

Content type
application/json
{
  • "message": "/regex/"
}

Restore MongoDB data using workspace metadata

Restores all MongoDB databases using provided workspace cloud credentials

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

Request Body schema: application/json
workspace_metadata
required
object

Workspace metadata containing cloud credentials

backup_file_name
string

Optional backup file name to restore from

Responses

Request samples

Content type
application/json
{
  • "workspace_metadata": { },
  • "backup_file_name": "string"
}

Response samples

Content type
application/json
{
  • "message": "/regex/"
}

INFERENCE_MICROSERVICE

Start Inference Microservice

Creates a new Inference Microservice job and starts the StatefulSet. Returns job_id for subsequent operations.

  • Creates a new unique job_id
  • Starts Inference Microservice StatefulSet microservice
  • Returns job_id for file uploads and inference requests
Authorizations:
bearer-token
path Parameters
org_name
required
string

Name of the organization

experiment_id
required
string

Experiment ID

Request Body schema: application/json
required
model_path
required
string

Path to the model

docker_image
string

Docker image for inference

gpu_type
string

GPU type required

num_gpus
integer

Number of GPUs required

Responses

Request samples

Content type
application/json
{
  • "model_path": "/workspace/model",
  • "docker_image": "nvcr.io/nvidia/vila-inference:latest",
  • "gpu_type": "H100",
  • "num_gpus": 1
}

Response samples

Content type
application/json
{
  • "job_id": "string",
  • "status": "string",
  • "message": "string"
}

Make Inference Microservice inference request

Sends inference request to a running Inference Microservice. This endpoint:

  • Validates the Inference Microservice is running
  • Processes prompts and images for Inference Microservice inference
  • Returns generated content from the Inference Microservice model
  • Supports both single and batch inference requests
Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

Experiment ID

job_id
required
string <uuid> <= 36 characters

Inference Microservice Job ID

Request Body schema: application/json
input
required
Array of strings

Base64-encoded images/videos with data URI format (data:image/jpeg;base64,...)

model
required
string

Model identifier (e.g. nvidia/nvdino-v2)

prompt
string
Default: ""

Text prompt for Inference Microservice inference

Responses

Request samples

Content type
application/json
{
  • "input": [
    ],
  • "model": "string",
  • "prompt": ""
}

Get Inference Microservice status

Returns the status of a Inference Microservice.

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

Experiment ID

job_id
required
string <uuid> <= 36 characters

Inference Microservice Job ID

Responses

Stop Inference Microservice

Stops a running Inference Microservice.

Authorizations:
bearer-token
path Parameters
org_name
required
string <= 255 characters ^[a-zA-Z0-9_-]+$

Org Name

experiment_id
required
string <uuid> <= 36 characters

Experiment ID

job_id
required
string <uuid> <= 36 characters

Inference Microservice Job ID to stop

Responses