Unverified50% confidenceFactExact time
OpenAI的o1/o3系列和Google的Gemini Thinking采用类似扩展思考的慢思考机制
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/20/2026
First Seen
Valid until: 10/18/2026
Sources
Related Claims
UnverifiedGoogle的Gemini 2.0 Flash Thinking和DeepSeek-R1等模型跟进了o1的推理时计算扩展技术路线73% similarUnverifiedGoogle 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71% similarVerifiedGemini 2.5具有思考模式（Thinking Mode），模型会先生成内部推理链再给出最终答案68% similarVerifiedGoogle Gemini在多模态理解上具有优势67% similarUnverifiedGoogle DeepMind在Gemini技术报告中明确指出Gemini系列从设计之初就是多模态原生的，而非在语言模型基础上后期添加视觉能力66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/566768API
curl https://kongchang.com/api/v1/knowledge/claims/566768MCP
get_claim(id=566768)