Verified80% confidenceFactTime unknown

GPT-4、Claude 3.5、DeepSeek等模型在HumanEval编程基准测试上的通过率已超过80%

15

Sources

80%

Confidence

Medium-term (~90 days)

Relevance

5/31/2026

First Seen

Valid until: 8/29/2026

Sources

零代码打造多平台内容分发助手：AI Coding实战

bilibili小弟姚安

Related Entities

GPT-4model Claude 3.5model DeepSeekmodel HumanEvalconcept

Related Claims

UnverifiedGPT-4和Claude 3在HumanEval代码基准测试上的pass@1通过率超过85%83% similar Unverified在标准化测试中，GPT-4的表现已超过90%的人类考生79% similar UnverifiedGPT-4级别的模型作为Judge时，其评判结果与人类专家的一致性超过80%77% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/5091

API

curl https://kongchang.com/api/v1/knowledge/claims/5091

MCP

get_claim(id=5091)