Trustible’s Model Transparency Ratings offers a comprehensive assessment of transparency disclosures of the top Large Language Models (LLMs). The Model Transparency Ratings assesses these GPAI LLM model families based on public documentation from sources like developers' websites, Github, and HuggingFace. Each model is evaluated across 30 distinct criteria in alignment with the EU AI Act GPAI transparency requirements and assigned overall scores. The ratings – inspired by other transparency rating reports such as financial credit ratings or ESG scores – offer an accessible format to help organizational AI leaders make informed, responsible decisions regarding model selection considering both ethical and legal risks.
For additional information about the Model Transparency Ratings, please read our blog post. Trustible has also published information on the methodology and evaluation criteria for the Model Transparency Ratings.
Each Model has a Model Rating Report that includes Overview information on the model along with the summary of Ratings, Transparency Assessment, and Source links. Model Cards also breakdown information by Basic Details, Policy, Model and Training, Data, and Evaluation. You can find additional information within the Trustible lightbulb for each data point.
