symflower/eval-dev-quality

Do an up-to-date leaderboard/dashboard for current models current evaluation

Opened this issue · 0 comments

Blog posts are nice, but it would be better to be up to date, and make the information available in a clearer way.

A good example is of course https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard. Maybe we can use the HuggingFace platform for the leaderboard/dashboard?