Reproduce “Keepace.ai: A Technical Report on a Large Language Model for Sports and Health” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Keepace.ai Team, Bowen Wu, Heng Wang, Jufeng Mi, Kang Liu, Kun Li, Long He, Manling Han and 7 more
Keepace.ai is a Chinese-language exercise-guidance model post-trained from Qwen3.5-27B for workout planning, sports-and-health questions, and personal workout-data analysis. It uses expert-designed rubrics to guide training and internal evaluation. Across 100 held-out Chinese prompts per task, Keepace.ai scored 92.88 overall versus 88.53 for Qwen3.5-397B-A17B, but ranked second on data analysis; these results are not real-world outcomes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.