Unverified50% confidenceBenchmarkExact time
GPT-5.6在知识工作、科学与健康领域测试全面领先Claude Opus及Opus 4.5
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/14/2026
First Seen
Valid until: 10/12/2026
Sources
GPT-5.6发布：三款模型对标Claude，ChatGPT Work超级应用全解析
bilibiliAI-seeker7/9/2026
Related Claims
UnverifiedSWE-1.7被宣称在编程智能水平上已接近GPT-5.5和Claude Opus的表现78% similarUnverified在网络安全测试排名中，Claude Opus 4.6最初领先，GPT 5.5发布后与之持平，GPT 5.5 Cyber一度领先，最终Claude新预览版重回榜首77% similarUnverified测试的两个项目此前已用Claude Opus 4.5加GPT-5.5的组合写过一轮76% similarVerifiedM2.5在编程任务上跑出了接近Claude Opus 4.6和GPT-5.2等旗舰级模型的表现75% similarUnverifiedGPT-5.5和Claude Opus 4.7在OpenRouter基准测试中占据智商最高位置，价格约5美元/百万Token74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/504395API
curl https://kongchang.com/api/v1/knowledge/claims/504395MCP
get_claim(id=504395)