OpenAI o-series
A family of reasoning-oriented models released by OpenAI that produce an internal chain-of-thought reasoning trace before delivering a final answer. These models are designed to perform well on complex reasoning tasks.
Core Facts
Reasoning models, represented by OpenAI o1, DeepSeek-R1, and QwQ, have chain-of-thought expansion as their core mechanism, generating a large number of intermediate reasoning steps before producing a final answer.
推理能力基于思维链（Chain-of-Thought, CoT）技术，模型在给出最终答案前生成内部逐步推导过程
Google's Gemini, Anthropic's Claude, and OpenAI's o-series models have all implemented different forms of extended thinking.
Timeline (last 90 days)
Google's Gemini, Anthropic's Claude, and OpenAI's o-series models have all implemented different forms of extended thinking.
In extended thinking mode, Claude counts thinking tokens and final output tokens against the same max_tokens budget, whereas OpenAI's o-series models have a separate budget cap for reasoning tokens.
The overthinking problem is not unique to Qwen3; reasoning models such as DeepSeek-R1 and the OpenAI o1 series all exhibit similar phenomena to varying degrees.
All Facts (6)
Reasoning models, represented by OpenAI o1, DeepSeek-R1, and QwQ, have chain-of-thought expansion as their core mechanism, generating a large number of intermediate reasoning steps before producing a final answer.
90%Verified推理能力基于思维链（Chain-of-Thought, CoT）技术，模型在给出最终答案前生成内部逐步推导过程
80%VerifiedGoogle's Gemini, Anthropic's Claude, and OpenAI's o-series models have all implemented different forms of extended thinking.
75%VerifiedReasoning models, represented by OpenAI's o series, DeepSeek-R1, and the Kimi K series, perform a chain-of-thought process before outputting, decomposing problems into multiple steps, performing self-verification, and deriving answers incrementally.
65%UnverifiedThe overthinking problem is not unique to Qwen3; reasoning models such as DeepSeek-R1 and the OpenAI o1 series all exhibit similar phenomena to varying degrees.
60%UnverifiedIn extended thinking mode, Claude counts thinking tokens and final output tokens against the same max_tokens budget, whereas OpenAI's o-series models have a separate budget cap for reasoning tokens.
50%