Browse and submit evaluations for CaselawQA benchmarks
Evaluate adversarial robustness using generative models
View and compare language model evaluations
Track, rank and evaluate open LLMs and chatbots
Evaluate model predictions with TruLens
Determine GPU requirements for large language models
Submit models for evaluation and view leaderboard
Generate leaderboard comparing DNA models
View and submit LLM benchmark evaluations
Convert and upload model files for Stable Diffusion
Calculate GPU requirements for running LLMs
Rank machines based on LLaMA 7B v2 benchmark results
Compare LLM performance across benchmarks
The CaselawQA leaderboard (WIP) is a platform designed for tracking and comparing the performance of AI models on the CaselawQA benchmark. It enables researchers and practitioners to evaluate and submit results for their models, fostering collaboration and progress in legal AI applications. The leaderboard is currently a work in progress, with ongoing updates and improvements being made to enhance its functionality and usability.
What is the CaselawQA benchmark?
The CaselawQA benchmark is a dataset and evaluation framework designed to assess the ability of AI models to answer legal questions based on case law.
How do I submit my model's results?
To submit your model's results, use the submission interface on the CaselawQA leaderboard. Follow the provided instructions to upload your results in the required format.
Is the leaderboard open to everyone?
Yes, the leaderboard is open to all researchers and developers who want to evaluate their models on the CaselawQA benchmark. No special access is required.