Track, rank and evaluate open LLMs and chatbots
Evaluate RAG systems with visual analytics
Explore and visualize diverse models
Evaluate reward models for math reasoning
View NSQL Scores for Models
Compare audio representation models using benchmark results
Visualize model performance on function calling tasks
Analyze model errors with interactive pages
Display leaderboard for earthquake intent classification models
Measure over-refusal in LLMs using OR-Bench
Merge machine learning models using a YAML configuration file
View LLM Performance Leaderboard
Determine GPU requirements for large language models
The Open LLM Leaderboard is a comprehensive tool designed to track, rank, and evaluate open-source Large Language Models (LLMs) and chatbots. It provides a transparent and standardized platform to compare models based on various benchmarks and metrics, helping developers, researchers, and users make informed decisions. By focusing on performance, efficiency, and capabilities, the Leaderboard serves as a go-to resource for understanding the evolution and advancements in the field of LLMs.
What metrics are used to rank LLMs? The Leaderboard uses a variety of metrics, including performance benchmarks, speed, memory usage, and specific task accuracy to ensure a holistic evaluation of each model.
Can I compare custom or non-listed models? Yes, the platform allows users to input custom models for comparison, providing flexibility for researchers and developers working on niche or proprietary LLMs.
How often is the Leaderboard updated? The Leaderboard is updated regularly to reflect new releases and improvements in existing models, ensuring users always have access to the latest information.