general leaderboards: 248 curated resources
general leaderboards
248 resources · page 3 of 8
Flames
by OpenAI45Lab
Flames GitHub repo: an adversarial Chinese benchmark of prompts on fairness, safety, morality, legality and data protection.
GEO-Bench
by ServiceNow
GEO-Bench GitHub repo from ServiceNow: benchmark for foundation models on Earth-observation classification and segmentation.
Global-MMLU
by Cohere Labs
Cohere Labs Global-MMLU dataset: MMLU translated into many languages with labels marking culturally sensitive questions.
GLUE
by GLUE team (NYU, University of Washington and DeepMind)
GLUE benchmark site: collection of natural-language-understanding tasks with a leaderboard for comparing models.
Habitat Challenge
Annual Habitat Challenge for embodied AI, with simulated navigation and object-rearrangement tasks, rules and leaderboards for agent evaluation.
HAE-RAE Bench
HAE-RAE Bench GitHub repo: a Korean benchmark on vocabulary, history, general knowledge and reading comprehension.
Hebrew LLM Leaderboard
by Open Hebrew LLM's Leaderboard
Hebrew LLM Leaderboard Hugging Face Space: model results on Hebrew language benchmarks.
HELM (Stanford CRFM)
Stanford CRFM HELM: holistic evaluation of language models across many scenarios and metrics.
HELMET
by Princeton Natural Language Processing
HELMET GitHub repo from Princeton: benchmark for long-context language models across seven application categories.
HF Find a Leaderboard
by OpenEvals
Hugging Face Find a Leaderboard Space: searchable directory of leaderboards hosted on the Hub.
Hugging Face LightEval
by Hugging Face
Hugging Face LightEval GitHub repo: toolkit for evaluating LLMs across many backends and tasks.
Humanity's Last Exam
by Center for AI Safety and Scale AI
Humanity's Last Exam: dataset of 2,500 expert-written questions across academic fields, with paper and results.
Humanloop
by Humanloop
Humanloop site, now a notice that the team joined Anthropic and the platform was shut down with a migration guide.
IberBench
by Universitat Politècnica de València
UPV repository record for IberBench, a benchmark for evaluating LLMs on Iberian languages such as Spanish, Catalan and Basque.
IFBench
by Valentina Pyatkin et al. (University of Washington)
IFBench paper on generalising verifiable instruction following, with out-of-distribution output constraints for evaluating LLMs.
INCLUDE
by Cohere Labs
Cohere Labs INCLUDE dataset: questions from local academic and professional exams in 44 languages for regional knowledge evaluation.
IndicGLUE
AI4Bharat IndicGLUE page: natural-language-understanding benchmark for Indian languages with tasks and datasets.
IndoMMLU
by IndoLEM
IndoMMLU dataset: questions from Indonesian school exams for evaluating language models on local knowledge.
IndoNLU
by IndoNLP
IndoNLU GitHub repo: Indonesian natural-language-understanding benchmark with tasks, pretrained models and starter code.
InfiniteBench
by OpenBMB
InfiniteBench GitHub repo: long-context tasks beyond 100K tokens for evaluating language models.
Inspect Evals
by Department of Business, Energy and Industrial Strategy
UK AISI inspect_evals GitHub repo: collection of evaluations implemented for the Inspect AI framework.
IrokoBench
Masakhane IrokoBench collection on Hugging Face: human-translated NLI, MMLU and math tasks in 16 African languages.
ISP RAS Trusted AI Research Center
by Ivannikov Institute for System Programming (ISP RAS)
ISP RAS research center for trustworthy AI: unit of the Russian Academy of Sciences' Institute for System Programming.
JGLUE
by Yahoo! JAPAN
JGLUE GitHub repo from Yahoo Japan: Japanese general language understanding benchmark built natively in Japanese.
JudgeBench
by ScalerLab
JudgeBench GitHub repo: benchmark for evaluating LLM-based judges on verifiable response-comparison tasks.
Kazakh LLM leaderboard
Kaz LLM Leaderboard Hugging Face Space: Kazakh-language benchmark results with model submissions.
KazMMLU
KazMMLU paper: a knowledge benchmark of Kazakhstan-specific questions in Kazakh and Russian for evaluating language models.
KLUE
KLUE GitHub repo: Korean natural-language-understanding benchmark with eight tasks and baseline code.
KMMLU
by Guijin Son et al. (Yonsei University)
KMMLU paper: a Korean multitask knowledge benchmark written natively in Korean rather than translated from English MMLU.
KMMLU-Redux and KMMLU-Pro
by Seokhee Hong et al. (LG AI Research)
Paper introducing KMMLU-Redux, a corrected version of KMMLU, and KMMLU-Pro, built from Korean professional licensing exams.
KMMLU-Redux / KMMLU-Pro
LG AI Research publication page for KMMLU-Redux and KMMLU-Pro, Korean benchmarks built from professional licensing exams.
La Leaderboard
La Leaderboard Hugging Face Space: evaluation of LLMs on Spanish and other languages of Spain and Latin America.