OAG-Bench is a comprehensive, multi-aspect, and fine-grained human-curated benchmark for academic graph mining, built on the Open Academic Graph (OAG). The benchmark covers 10 tasks, 20 datasets, 70+ baseline methods, and 120+ experimental results. It aims to address the limitations of existing academic graphs and benchmarks by providing detailed annotations, data preprocessing codes, algorithm implementations, and standardized evaluation protocols. OAG-Bench includes tasks such as author name disambiguation, scholar profiling, paper source tracing, and academic question answering. The paper also introduces the Open Academic Graph Challenge (OAG-Challenge) to encourage community participation and sharing. The benchmark is designed to facilitate the development and evaluation of advanced algorithms in academic graph mining, making it a valuable resource for researchers and practitioners.OAG-Bench is a comprehensive, multi-aspect, and fine-grained human-curated benchmark for academic graph mining, built on the Open Academic Graph (OAG). The benchmark covers 10 tasks, 20 datasets, 70+ baseline methods, and 120+ experimental results. It aims to address the limitations of existing academic graphs and benchmarks by providing detailed annotations, data preprocessing codes, algorithm implementations, and standardized evaluation protocols. OAG-Bench includes tasks such as author name disambiguation, scholar profiling, paper source tracing, and academic question answering. The paper also introduces the Open Academic Graph Challenge (OAG-Challenge) to encourage community participation and sharing. The benchmark is designed to facilitate the development and evaluation of advanced algorithms in academic graph mining, making it a valuable resource for researchers and practitioners.