Open LLM Leaderboard
huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard
Hugging Face's primary benchmark leaderboard for open-weight language models, evaluating hundreds of models across standardized reasoning, knowledge, and coding benchmarks.
Why it is useful
The most comprehensive and consistently updated ranking of open-weight models, covering everything from 1B parameter models to 70B+ ones across multiple benchmark categories. When choosing a base model for fine-tuning or local deployment, the leaderboard lets you compare capabilities at specific parameter counts and find the best model that fits your hardware constraints.
More in Research & Data
Other research & data you might not have heard of.
- Stanford Network Analysis ProjectCollection of large network datasets — social graphs, communication patterns, web graphs, citation networks — for graph and network science research.
- WikidataFree, structured knowledge base underlying Wikipedia's infoboxes — every fact is machine-readable and directly queryable.
- KaggleCommunity platform for datasets, machine learning competitions, and collaborative notebooks.
- iNaturalistA community platform for recording and identifying plants, animals and fungi, where photos are reviewed by other naturalists and feed into open biodiversity data.
- EurostatThe statistical office of the European Union, providing free access to harmonized statistics across all EU member states.
- PubMedThe National Library of Medicine's free database of over 35 million citations for biomedical literature, with full-text links where available.