Arena-Hard-Auto selects 500 prompts from 200,000 real LMArena user conversations. Answers are scored automatically with a strong model as judge, which is reported to correlate closely with human preference results.
Size
500 prompts in the source; none imported.
License
apache-2.0 (declared on the Hugging Face dataset card)
IFEval measures whether a model follows verifiable instructions -- word counts, formats, forbidden words -- rather than whether an answer is good. Every constraint is checked programmatically.
Full text541 tasks
Size
541 problems imported from the default / train selection (the full split; the source names its only split train).
MT-Bench is a set of 80 two-turn questions across eight categories, published with the study that established LLM-as-a-judge evaluation. It looks at multi-turn consistency and instruction following.
Metric
average judge score over the eight categories, reported per turn
Size
80 questions in the source; none imported.
License
apache-2.0 (declared on the Hugging Face dataset card)