Unverified50% confidenceBenchmarkExact time

OpenAI于2021年发布Codex技术报告，该模型在HumanEval基准测试上的首次通过率达28.8%，经迭代改进后超过72%

1

Sources

50%

Confidence

Long-term

Relevance

7/8/2026

First Seen

Sources

Codex接入外置模型完整指南：两种方法解锁全部能力

bilibili长青来了奥6/8/2026

Related Claims

VerifiedHumanEval由OpenAI于2021年提出，考察函数级代码补全68% similar Unverified规模定律由OpenAI研究人员于2020年系统量化，后经Chinchilla论文（2022）修正66% similar VerifiedThe earlier Codex API was released by OpenAI in 2021.65% similar UnverifiedOpenAI在2023年开源了其Evals框架，将评估分为基于事实的精确匹配、基于模型的评判、以及基于人类标注的评估三大类65% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/210563

API

curl https://kongchang.com/api/v1/knowledge/claims/210563

MCP

get_claim(id=210563)