A series of new models have been added to the leaderboard, showcasing a range of performance metrics across various categories. Notably, one model achieved an overall rank of 48, ranking first in the 13B category with an average score of 40.69, an IFEval score of 73.05, and a BBH score of 49.51. Another model ranked 44 in the 13B category with an average score of 41.22 and an IFEval score of 72.11. Other models included those ranked 127 with an average of 37.48 and an IFEval of 53.61, and 606 with an average of 28.96 and an IFEval of 73.71. Additionally, Aya Expanse has been recognized as the best open-weights model on @scale's private multilingual protocol, outperforming both proprietary and larger models in certain languages, indicating strong multilingual capabilities.