[Question] Are no models being evaluated due to the lighteval/MATH-Hard dataset being unavailable?

#1071
by T145 - opened

I noticed the backlog, tried to run tests myself, and came across this issue:
https://github.com/huggingface/lighteval/issues/494

Is that why deepseek v3 is not in the list? I want to see this 685B monster crushing it.

Open LLM Leaderboard org

Models still have access to the dataset for evaluation - cluster is full however, hence evaluations go more slowly :)

clefourrier changed discussion status to closed

Sign up or log in to comment