Unverified60% confidenceFactTime unknown
BrowseComp是OpenAI发布的网页浏览与信息检索能力基准测试
2
Sources
60%
Confidence
Long-term
Relevance
5/31/2026
First Seen
Sources
Gemini 3.1 Pro深度横评：ARC-AGI-2得分77.1%，真的最强吗？
bilibili锋芒AI
Related Entities
Related Claims
VerifiedBrowseComp是OpenAI提出的代理搜索基准，考察模型在真实网络环境中完成多步信息检索与整合的能力83% similarUnverifiedOpenAI的Evals框架和PromptFlow提供了Prompt的自动化测试能力73% similarUnverifiedOpenAI 的 Sites 潜在优势包括自有模型的语义理解能力、ChatGPT庞大用户基础及潜在多模态输入能力71% similarUnverifiedopen-code-review同时兼容OpenAI和Anthropic的API协议71% similarUnverified任务集中在网页操作时，OpenAI Operator或结合DOM解析的开源框架往往能提供更好的成功率71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/12521API
curl https://kongchang.com/api/v1/knowledge/claims/12521MCP
get_claim(id=12521)