Humanity's Last Exam (HLE) is a
language model benchmark consisting of 2,500 questions across a broad range of subjects. It was created jointly by the
Center for AI Safety and
Scale AI.
Creation
Stanford HAI's AI Index 2025 Annual Report cites Humanity's Last Exam as one of the "more challenging benchmarks" developed in response to the popular AI benchmarks having reached "saturation".
The test has been described as the brainchild of
Dan Hendrycks, a machine learning researcher and the director of the
Center for AI Safety, who stated that he was inspired to create the test after a conversation with
Elon Musk
Elon Reeve Musk ( ; born June 28, 1971) is a businessman. He is known for his leadership of Tesla, SpaceX, X (formerly Twitter), and the Department of Government Efficiency (DOGE). Musk has been considered the wealthiest person in th ...
, who thought the existing
language model benchmarks, such as the
MMLU, were too easy. Hendrycks worked with
Scale AI to compile the questions.
The questions were
crowdsourced from subject matter experts from various institutions across the world.
The questions were first filtered by the leading AI models; if the models failed to answer the question or did worse than random guessing on the multiple-choice questions, they were reviewed by human experts in two rounds and approved for inclusion in the dataset. The submitters of the top-rated questions were given prize money from a pool of 500,000
U.S. dollars—$5000 for each of the top 50 questions and $500 for the next 500. After the initial release, a "community feedback bug bounty program" was opened to "identify and remove major errors in the dataset".
Composition
The benchmark consists of 2,500 questions in the publicly released set. The paper classifies the questions into the following broad subjects: mathematics (41%), physics (9%), biology/medicine (11%), humanities/social science (9%), computer science/artificial intelligence (10%), engineering (4%), chemistry (7%), and other (9%). Around 14% of the questions require the ability to understand both text and images, i.e.,
multi-modality. 24% of the questions are multiple-choice; the rest are short-answer, exact-match questions. A private set is also maintained to test for benchmark
overfitting
In mathematical modeling, overfitting is "the production of an analysis that corresponds too closely or exactly to a particular set of data, and may therefore fail to fit to additional data or predict future observations reliably". An overfi ...
.
An example question:
Results
References
Large language models
2025 in artificial intelligence
External links
Humanity's Last Examat the
Center for AI Safety.
Humanity's Last Examat
Scale AI.
{{Artificial intelligence navbox