nguyen599/AstralBench download history
nguyen599/AstralBench is a text generation dataset on the Hugging Face Hub. In the last 30 days it was downloaded 28 times (11 in the last 7 days), and 1,198 times in total. It ranks #289,218 among datasets by monthly downloads.
AstralBench is a carefully curated subset of 50 high-quality problems, selected for benchmarking model performance. It covers diverse mathematical topics and difficulty levels, with current model performance ranging from 5% to 30% accuracy. Source of AstralBench Seed data: IMO AnswerBench
Open nguyen599/AstralBench on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.