What AI says about MixEval

MixEval is a benchmark that evaluates large language models by strategically mixing off-the-shelf benchmarks to bridge real-world user queries with ground-truth-based evaluation. It achieves a 0.96 ranking correlation with Chatbot Arena while running at 6% of the time and cost of MMLU, and supports dynamic updates to prevent contamination.