Unverified60% confidenceFactTime unknown
Google Gemini在最新的数学推理和编程基准测试中已经超越了OpenAI O3、Claude 4和DeepSeek R1，稳居榜首
1
Sources
60%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026
Sources
Google AI Studio使用教程：Gemini实战指南从入门到精通
bilibili出海学长Ethan
Related Entities
Related Claims
UnverifiedGemini 3在SWE-bench基准测试中智能体场景下的得分远超同类模型79% similarUnverified一位技术博主对Google的Gemini 3.1 Pro和Anthropic的Opus 4.6进行了为期两天的前端编程能力全方位测试77% similarUnverified五项基准测试综合结果为技术性平局，Claude 4.5在编程实战领先，Gemini 3 Pro在知识推理领先76% similarUnverified在研究生级推理和Agent工具使用基准测试中，Opus 4基本处于领先地位，但Gemini 2.5 Pro在部分项目上与之非常接近75% similarUnverifiedQwen3.5-Omni多项关键指标超过Google的Gemini 3.1 Pro74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/23825API
curl https://kongchang.com/api/v1/knowledge/claims/23825MCP
get_claim(id=23825)