Darwin tops 12 of 57 Hugging Face official benchmark leaderboards, per Maeil Business analysis

Maeil Business reported on November 17, 2026 that Hugging Face's official benchmark set includes 63 benchmarks, 57 leaderboards, 1,458 recorded results and 580 models. Bidraft's Darwin model ranked first on 12 of these leaderboards, ahead of Moonshot AI with 7, Z.ai with 6, DeepSeek with 3 and Alibaba Qwen with 3. Bidraft placed in the top three on all 15 leaderboards where it submitted results, and the author notes the counts can be reproduced by anyone with a short Python script against the public API. The article also flags that ranking by first-place count favors organizations that enter more leaderboards, so per-entry win rates are included as well. By country, Chinese organizations hold 34 first-place finishes, South Korea 14 and the United States 7.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.


Discussion (0)
Log in to join the discussion and vote.
Log in