Using User Environment APIs to Manage LLMs - Teradata VantageCloud Lake

Lake - Analyze Your Data with ClearScape Analytics™

Deployment
VantageCloud
Edition
Lake
Product
Teradata VantageCloud Lake
Release Number
Published
February 2025
ft:locale
en-US
ft:lastEdition
2026-02-20
dita:mapPath
tcl1683670667798.ditamap
dita:ditavalPath
pny1626732985837.ditaval
dita:id
tcl1683670667798

After the LLM directories are compressed into zip files, use these user environment model APIs to install, uninstall, and list LLMs.

To manage deep learning and other models in file format, use the APIs install_files(), files, and remove_files().

install_model

Create a Python_3.10 environment with given name and description in the database.

>>> env = create_env("test_env", "python_3.10", "Test environment")
User environment 'test_env' created.
>>> model = 'large_model.zip'

Install the model in the 'test_env' environment.

>>> env.install_model(model_path = model)
Request for install_model is completed successfully.
                               Claim Id  File/Libs/Model    Method Name               Stage             Timestamp  Additional Details
0  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model  Endpoint Generated  2025-07-09T09:22:30Z
1  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model       File Uploaded  2025-07-09T09:22:31Z
2  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model      File Installed  2025-07-09T09:22:52Z
>>> env.install_model(model_name="google-bert/bert-base-uncased")
Request for install_model is completed successfully.
Model 'google-bert/bert-base-uncased' installed successfully in the remote user environment 'test_env'.
>>> env.install_model(model_name="mistralai/Mistral-7B-v0.1", api_key="abcd1234", asynchronous=True)
Model installation is initiated. Check the status using <UserEnv_obj>.status() with the claim id 'ac284706-0b72-4b83-8add-3cff632747f4'.

Check status using claim-id.

>>> env.status(claim_id)
                                 Claim Id    File/Libs/Model                 Method Name    Stage                         Timestamp      Additional Details
0    ac284706-0b72-4b83-8add-3cff632747f4    mistralai/Mistral-7B-v0.1    install_model    Started            2025-07-30T03:58:09Z    Begin downloading model named mistralai/Mistral-7B...
1    ac284706-0b72-4b83-8add-3cff632747f4    mistralai/Mistral-7B-v0.1    install_model    ModelInstalled     2025-07-30T03:58:28Z    Model installed successfully
# Create a Python_3.10 environment with given name and description in Vantage.
>>> env = create_env("test_env", "python_3.10", "Test environment")
User environment 'test_env' created.
>>> model = 'large_model.zip'
# Install the model in the 'test_env' environment.
>>> env.install_model(model_path = model)
Request for install_model is completed successfully.
                               Claim Id  File/Libs/Model    Method Name               Stage             Timestamp  Additional Details
0  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model  Endpoint Generated  2023-11-09T09:21:24Z
1  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model       File Uploaded  2023-11-09T09:22:28Z
2  766423af-7cc4-46db-8ee2-b6b5ded299c6  large_model.zip  install_model      File Installed  2023-11-09T09:22:30Z

list models

# List models.
>>> env.models
         Model  Size             Timestamp
0  large_model  6144  2023-11-09T09:22:30Z

uninstall_model

# Uninstall model from remote user environment.
>>> env.uninstall_model('large_model')
Model 'large_model' uninstalled successfully from the remote user environment 'test_env'.
True