HomeAI Tutorial

truthlens-ai

Public

Real-Time Truthfulness Score for LLM Output. Evaluates AI-generated text for hallucinations by checking claims against external evidence sources. Features automatic claim extraction, evidence retrieval from Wikipedia, truthfulness scoring (0-100%), support for OpenAI and Anthropic models, Streamlit UI, and FastAPI backend.

Creat2025-11-23T00:11:46
Update2025-11-24T01:08:50
0
Stars
0
Stars Increase

Related projects