Wiki
Tags
About
FAQ
RSS
AIPOST

HumanEval

HumanEval is a benchmark released by OpenAI for evaluating how well AI models generate Python code.

1 article
Last mentioned

HumanEval is a benchmark OpenAI released in 2021. It gives a model a function description, asks it to write the Python function, and scores the answer by whether it passes unit tests. It consists of 164 problems and is widely used to compare the coding ability of large language models.

This entry is based on AIPOST articles and widely known facts. If something is wrong, please send us a correction request.

Articles covering this entry

HumanEval Remix, 100 problems 75 passed (75%), 5 left unanswered

Read article

Related entries

Entries that often appear in the same articles.

Browse the wikiReport & corrections

prompt

A prompt is an instruction, question or contextual input that tells an AI model what task to perform.

77 articles

token

A token is a unit of text processed by an AI model; it does not necessarily equal one word or character.

62 articles

OpenAI

OpenAI is a US AI research and product company known for its GPT models and ChatGPT.

60 articles
AIPOST
© 2026 AIPOST. All rights reserved.

AIPOST is an AI publication covering practical AI, AI security, performance, startups, health, ethics and industry news. No account is needed, and our privacy policy explains how we handle personal information.