Unverified75% confidenceFactTime unknown

GPT-5发布后跑分数据显示其几乎全面超越Claude等所有模型

1

Sources

75%

Confidence

Medium-term (~90 days)

Relevance

6/1/2026

First Seen

Valid until: 8/30/2026

Sources

GPT-5 vs Claude 4编程实测：全栈项目深度对比结果出炉

bilibili不正经的前端啊

Related Entities

GPT-5model Claudemodel

Related Claims

Unverified据实测评价，GLM-5.2的综合表现已达到GPT-5.x和Claude系列旗舰模型的水平线84% similar VerifiedGPT-5.5在多项能力上全面超越Claude 4.7和Gemini 3.1 Pro等模型79% similar UnverifiedGPT-5.5发布才不过几周，其表现远超Claude Opus 4.778% similar UnverifiedGPT 5.5在TerminalBench 2.0基准测试中以87%的成绩大幅领先，超过了Anthropic尚未发布的Mythos模型77% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/18620

API

curl https://kongchang.com/api/v1/knowledge/claims/18620

MCP

get_claim(id=18620)