Unverified50% confidenceBenchmarkExact time
HumanEval由OpenAI自研，专门测试代码生成的功能正确率
1
Sources
50%
Confidence
Long-term
Relevance
7/17/2026
First Seen
Sources
Related Claims
VerifiedHumanEval是由OpenAI于2021年发布的代码生成基准测试，包含164个手写的Python编程问题80% similarUnverified代码生成领域的模型评估如HumanEval、SWE-bench格外强调功能正确性而非流畅度75% similarUnverified代码大语言模型采用HumanEval、MBPP等可执行基准评估，即直接运行生成代码并检验输出是否正确74% similarUnverifiedHumanEval专注代码生成能力，以164道编程题检验模型的逻辑推理与代码正确性73% similarUnverifiedHumanEval基准测试由OpenAI于2021年随Codex模型论文一同发布，包含164道编程题，Pass@1指一次生成即通过全部测试用例的比例73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/547586API
curl https://kongchang.com/api/v1/knowledge/claims/547586MCP
get_claim(id=547586)