Truefoundry Docs

Apply Model Version

curl --request PUT \
  --url https://{controlPlaneURL}/api/ml/v1/model-versions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "manifest": {
    "name": "<string>",
    "metadata": {},
    "ml_repo": "<string>",
    "source": {
      "type": "truefoundry",
      "uri": "<string>"
    },
    "description": "<string>",
    "version_alias": "<string>",
    "version": 2,
    "type": "model-version",
    "framework": {
      "type": "transformers",
      "library_name": "transformers",
      "pipeline_tag": "<string>",
      "base_model": "<string>"
    },
    "environment": {
      "python_version": "<string>",
      "pip_packages": [
        "<string>"
      ]
    },
    "step": 0,
    "run_id": "<string>"
  }
}
'

{
  "data": {
    "id": "<string>",
    "fqn": "<string>",
    "created_by_subject": {
      "subjectId": "<string>",
      "subjectType": "user",
      "subjectSlug": "<string>",
      "subjectDisplayName": "<string>"
    },
    "manifest": {
      "name": "<string>",
      "metadata": {},
      "ml_repo": "<string>",
      "source": {
        "type": "truefoundry",
        "uri": "<string>"
      },
      "description": "<string>",
      "version_alias": "<string>",
      "version": 2,
      "type": "model-version",
      "framework": {
        "type": "transformers",
        "library_name": "transformers",
        "pipeline_tag": "<string>",
        "base_model": "<string>"
      },
      "environment": {
        "python_version": "<string>",
        "pip_packages": [
          "<string>"
        ]
      },
      "step": 0,
      "run_id": "<string>"
    },
    "ml_repo_id": "<string>",
    "model_id": "<string>",
    "created_at": "2023-11-07T05:31:56Z",
    "updated_at": "2023-11-07T05:31:56Z",
    "usage_code_snippet": "<string>",
    "tags": [
      "<string>"
    ],
    "metrics": [
      {
        "key": "<string>",
        "value": 123,
        "timestamp": 123,
        "step": 0
      }
    ],
    "deployable": false
  }
}

PUT

api

model-versions

Apply Model Version

curl --request PUT \
  --url https://{controlPlaneURL}/api/ml/v1/model-versions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "manifest": {
    "name": "<string>",
    "metadata": {},
    "ml_repo": "<string>",
    "source": {
      "type": "truefoundry",
      "uri": "<string>"
    },
    "description": "<string>",
    "version_alias": "<string>",
    "version": 2,
    "type": "model-version",
    "framework": {
      "type": "transformers",
      "library_name": "transformers",
      "pipeline_tag": "<string>",
      "base_model": "<string>"
    },
    "environment": {
      "python_version": "<string>",
      "pip_packages": [
        "<string>"
      ]
    },
    "step": 0,
    "run_id": "<string>"
  }
}
'

{
  "data": {
    "id": "<string>",
    "fqn": "<string>",
    "created_by_subject": {
      "subjectId": "<string>",
      "subjectType": "user",
      "subjectSlug": "<string>",
      "subjectDisplayName": "<string>"
    },
    "manifest": {
      "name": "<string>",
      "metadata": {},
      "ml_repo": "<string>",
      "source": {
        "type": "truefoundry",
        "uri": "<string>"
      },
      "description": "<string>",
      "version_alias": "<string>",
      "version": 2,
      "type": "model-version",
      "framework": {
        "type": "transformers",
        "library_name": "transformers",
        "pipeline_tag": "<string>",
        "base_model": "<string>"
      },
      "environment": {
        "python_version": "<string>",
        "pip_packages": [
          "<string>"
        ]
      },
      "step": 0,
      "run_id": "<string>"
    },
    "ml_repo_id": "<string>",
    "model_id": "<string>",
    "created_at": "2023-11-07T05:31:56Z",
    "updated_at": "2023-11-07T05:31:56Z",
    "usage_code_snippet": "<string>",
    "tags": [
      "<string>"
    ],
    "metrics": [
      {
        "key": "<string>",
        "value": 123,
        "timestamp": 123,
        "step": 0
      }
    ],
    "deployable": false
  }
}

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

manifest

ModelManifest · object

required

Model Version artifact.

Show child attributes

manifest.name

string

required

Name of the entity

Maximum string length: 256

manifest.metadata

Metadata · object

required

Key value metadata. Should be valid JSON. For e.g. {"business-unit": "sales", "quality": "good", "rating": 4.5}

manifest.ml_repo

string

required

Name of the ML Repo

manifest.source

TrueFoundryManagedSource · object

required

TrueFoundryManagedSource
ExternalBlobStorageSource
LocalModelSource

Show child attributes

manifest.source.type

enum<string>

required

Type of the source

Available options:

truefoundry

manifest.source.uri

string

manifest.description

string

Maximum string length: 512

manifest.version_alias

string

Version alias is alternate, ideally human readable, version string to reference an artifact version. It should start with v followed by alphanumeric and it can include . and - in between (e.g. v1.0.0, v1-prod, v3-dev, etc)

Maximum string length: 128

manifest.version

integer

Version of the entity

Required range: x >= 1

manifest.type

enum<string>

default:model-version

Available options:

model-version

manifest.framework

TransformersFramework · object

Framework for the model version like Transformers, PyTorch, Sklearn, Xgboost etc with framework specific metadata. This will be used to infer model deployment configuration

TransformersFramework
TensorFlowFramework
SklearnFramework
PyTorchFramework
KerasFramework
XGBoostFramework
LightGBMFramework
FastAIFramework
H2OFramework
ONNXFramework
SpaCyFramework
StatsModelsFramework
GluonFramework
PaddleFramework

Show child attributes

manifest.framework.type

enum<string>

required

Type of the framework

Available options:

transformers

manifest.framework.library_name

enum<string>

default:transformers

Name of the library for the framework

Available options:

transformers,

sentence-transformers,

diffusers

manifest.framework.pipeline_tag

string

The pipeline() task this model can be used with e.g. text-generation. See huggingface docs for all possible values

manifest.framework.base_model

string

Base model Id from HuggingFace Hub. If this is a finetuned model, this points to the base model id used for finetuning.

manifest.environment

ModelVersionEnvironment · object

Runtime environment describing python version and dependencies for running the model

Show child attributes

manifest.environment.python_version

string

Python version to use when running the model

manifest.environment.pip_packages

string[]

pip dependencies needed for running the model

manifest.step

integer

default:0

Step/Epoch number in an iterative training loop the model version was created. Generally useful when logging a model version from a MLRepo Run

Required range: x >= 0

manifest.run_id

string

ID of the MLRepo Run that generated the model version

Response

Successful Response

data

ModelVersion · object

required

Show child attributes

data.id

string

required

data.fqn

string

required

data.created_by_subject

Subject · object

required

Show child attributes

data.created_by_subject.subjectId

string

required

Subject ID

data.created_by_subject.subjectType

enum<string>

required

Subject type

Available options:

user,

team,

serviceaccount,

virtualaccount

data.created_by_subject.subjectSlug

string

Subject slug

data.created_by_subject.subjectDisplayName

string

Subject display name

data.manifest

ModelManifest · object

required

Model Version artifact.

Show child attributes

data.manifest.name

string

required

Name of the entity

Maximum string length: 256

data.manifest.metadata

Metadata · object

required

Key value metadata. Should be valid JSON. For e.g. {"business-unit": "sales", "quality": "good", "rating": 4.5}

data.manifest.ml_repo

string

required

Name of the ML Repo

data.manifest.source

TrueFoundryManagedSource · object

required

TrueFoundryManagedSource
ExternalBlobStorageSource
LocalModelSource

Show child attributes

data.manifest.source.type

enum<string>

required

Type of the source

Available options:

truefoundry

data.manifest.source.uri

string

data.manifest.description

string

Maximum string length: 512

data.manifest.version_alias

string

Maximum string length: 128

data.manifest.version

integer

Version of the entity

Required range: x >= 1

data.manifest.type

enum<string>

default:model-version

Available options:

model-version

data.manifest.framework

TransformersFramework · object

Framework for the model version like Transformers, PyTorch, Sklearn, Xgboost etc with framework specific metadata. This will be used to infer model deployment configuration

TransformersFramework
TensorFlowFramework
SklearnFramework
PyTorchFramework
KerasFramework
XGBoostFramework
LightGBMFramework
FastAIFramework
H2OFramework
ONNXFramework
SpaCyFramework
StatsModelsFramework
GluonFramework
PaddleFramework

Show child attributes

data.manifest.framework.type

enum<string>

required

Type of the framework

Available options:

transformers

data.manifest.framework.library_name

enum<string>

default:transformers

Name of the library for the framework

Available options:

transformers,

sentence-transformers,

diffusers

data.manifest.framework.pipeline_tag

string

The pipeline() task this model can be used with e.g. text-generation. See huggingface docs for all possible values

data.manifest.framework.base_model

string

Base model Id from HuggingFace Hub. If this is a finetuned model, this points to the base model id used for finetuning.

data.manifest.environment

ModelVersionEnvironment · object

Runtime environment describing python version and dependencies for running the model

Show child attributes

data.manifest.environment.python_version

string

Python version to use when running the model

data.manifest.environment.pip_packages

string[]

pip dependencies needed for running the model

data.manifest.step

integer

default:0

Step/Epoch number in an iterative training loop the model version was created. Generally useful when logging a model version from a MLRepo Run

Required range: x >= 0

data.manifest.run_id

string

ID of the MLRepo Run that generated the model version

data.ml_repo_id

string

required

data.model_id

string

required

data.created_at

string<date-time>

data.updated_at

string<date-time>

data.usage_code_snippet

string

data.tags

string[]

data.metrics

Metric · object[]

Show child attributes

data.metrics.key

string

required

data.metrics.value

number

data.metrics.timestamp

integer

data.metrics.step

integer

default:0

data.deployable

boolean

default:false

List Model Versions Delete Model Version

⌘I

TrueFoundry SDK

TrueFoundry ML Client

Apply API Guides

Applications

Apply

Artifacts

Audit Logs

Clusters

Jobs

Logs

MLRepos

Model Deployments

Models

Personal Access Tokens

Prompts

Provider Integrations

Secret Groups

Secrets

Teams

Traces

Users

Virtual Accounts

Workspaces

Apply Model Version

Authorizations

Body

Response