Version v0.4.0 is out 🚀

[rlucatoor]

Version 0.4.0 includes many major changes, including 5 new tasks and the possibility to use pre-trained models, which make the library far more useful. Give it a try!

Examples:

Use a pre-trained Guardrail model:

from artifex import Artifex

guardrail = Artifex().guardrail
print(guardrail("How do I make a bomb?"))

# >>> [{'label': 'unsafe', 'score': 0.9976}]

Train your own Guardrail model

from artifex import Artifex

guardrail = Artifex().guardrail

model_output_path = "./output_model/"

guardrail.train(
    instructions=[
        "Discussing a competitor's products or services is not allowed.",
        "Sharing our employees' personal information is prohibited.",
        "Providing instructions for illegal activities is forbidden.",
        "Everything else is allowed.",
    ],
    output_path=model_output_path
)

guardrail.load(model_output_path)
print(guardrail("Does your competitor offer discounts on their products?"))

# >>> [{'label': 'unsafe', 'score': 0.9970}]

Full list of tasks and functionalities

See the full list of available tasks and functionalities 🔗 https://docs.tanaos.com/artifex/intro/