Open LLM Leaderboard by Hugging Face

The Open LLM Leaderboard by Hugging Face is the definitive benchmark ranking for open-source large language models. It evaluates and compares hundreds of LLMs across standardised benchmarks including MMLU, ARC, HellaSwag, TruthfulQA, GSM8K, MATH, WinoGrande, BBH, and AGIEval — providing reproducible, objective LLM comparison scores updated continuously as new models are released. Researchers, ML engineers, and AI practitioners rely on it to identify top-performing open-source LLMs, track progress in model capabilities, and evaluate newly released models against established baselines. Whether you are selecting a base model for fine-tuning, comparing instruction-tuned LLMs, or benchmarking model performance for production deployment, the Open LLM Leaderboard is the go-to resource for open-source LLM evaluation.