Unverified75% confidenceFactTime unknown
在15款大模型开发B站首页的测试中，ChatGPT 5.4以82分的总分获得全场第一
1
Sources
75%
Confidence
Medium-term (~90 days)
Relevance
6/2/2026
First Seen
Valid until: 8/31/2026
Sources
实测15款大模型开发B站首页：GPT登顶，国产模型差距明显
bilibili圣徒城的小诺
Related Entities
Related Claims
UnverifiedMTP-LX 4bit版本在理财规划推理任务上经GPT-5.5 Thinking评分为55分（满分100），而同题GPT-5.5 Pro得分82分68% similarUnverified在Artificial Analysis第三方评测的编码指数（推理模式）上，GPT-5 Mini High的得分比Haiku 4.5高出9个点67% similarUnverifiedGLM 5.2在Terminal Bench测试中得分81分，GLM 5.1得分63.5分，GPT 5.5得分84分，Opus 4.8得分85分65% similarUnverifiedTerminal Bench 2.1测试中GPT-5.6 Ultra得分接近92%，Fable 5为88%，考虑误差范围基本持平63% similarUnverifiedGLM-5在SWE-Bench Verified基准测试中得分77.8%，位于开源第一梯队63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/38125API
curl https://kongchang.com/api/v1/knowledge/claims/38125MCP
get_claim(id=38125)