Unverified50% confidenceFactExact time
HumanEval由OpenAI设计，包含164道代码补全题目，考察模型将自然语言转化为可运行代码的能力
1
Sources
50%
Confidence
Long-term
Relevance
7/21/2026
First Seen
Sources
Related Claims
VerifiedHumanEval was proposed by OpenAI and contains 164 Python programming problems, primarily testing function-level code generation capability85% similarUnverifiedOpenAI once conducted internal experiments with Codex, with the goal of having humans write no code by hand and only be responsible for architectural design, while leaving all coding to agents.76% similarUnverified2020年OpenAI发布的Codex模型首次证明了大语言模型在复杂代码任务上的能力72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/579928API
curl https://kongchang.com/api/v1/knowledge/claims/579928MCP
get_claim(id=579928)