Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
TruthTorchLMgithub.com · Low evidenceTruthTorchLM is a Python package for assessing and predicting truthfulness in large language model outputs, featuring over 30 methods including Google search checks and uncertainty estimation. It integrates with Huggingface and LiteLLM, supports calibration and evaluation of truth methods, and handles both short-form QA and long-form text generation.
Parse Score