Welcome to OpenGVL! ✨

We're passionate about fostering the development of open-source Vision-Language Models (VLMs) for the challenging and exciting task of Generative Value Learning (GVL).

Our core mission is to provide a transparent and up-to-date leaderboard 🏆, showcasing the performance of various open-source VLMs on robotic benchmarks. We believe in making VLM evaluation accessible and clear for everyone.

What's Next? 🚀

We're constantly working to expand our leaderboard with even more cutting-edge open-source VLMs. Plus, our evaluation code will be released shortly, empowering you to dive deeper into our methodologies. Stay tuned for exciting updates!