Verified70% confidenceFactTime unknown
Gemini 2.5 features a Thinking Mode in which the model generates an internal reasoning chain before producing the final answer.
4
Sources
70%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGemini's model selector offers three modes — Fast, Thinking, and Pro — corresponding to everyday miscellaneous tasks, decisions requiring deep thinking, and heavy coding scenarios, respectively.78% similarUnverifiedGemini 2.5引入了原生的思维链（Chain-of-Thought）支持78% similarUnverifiedGemini 3.8 TTS's native multi-speaker support enables the model to perceive the entire conversation context within a single inference pass, eliminating the need for post-processing concatenation.76% similarUnverifiedGemini's thinking level can be switched from "Standard" to "Extended"; Extended mode performs more internal reasoning steps before giving an answer (similar to Chain-of-Thought), which helps with complex problems such as mathematical derivations and logic puzzles, but accelerates quota consumption.76% similarVerifiedThe Extended Thinking version of Gemini 3.8 Live incorporates a more thorough thinking process before voice responses, making it suitable for complex reasoning scenarios, while the standard version places greater emphasis on response speed.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/29456API
curl https://kongchang.com/api/v1/knowledge/claims/29456MCP
get_claim(id=29456)