This repository contains examples of using the Contextual API's in a standalone fashion.
LMUnit is a specialized model developed by Contextual AI for evaluating LLM response quality through natural language unit testing. It achieves state-of-the-art performance in assessing generative AI outputs by providing a structured method to test responses.
Key features:
- Natural language unit tests for LLM response evaluation
- State-of-the-art performance in quality assessment
- Structured testing methodology
- API access available (free access can be requested)
See the LMUnit example for a detailed walkthrough using financial services as a case study.
Generate is Contextual AI's Grounded Language Model.
See the Generate example for a demonstration of using the Generate API to power text generation applications.