Verified65% confidenceFactExact time
The number of internal reasoning tokens in OpenAI's o1/o3 series models is often several times to dozens of times that of the final output.
3
Sources
65%
Confidence
Long-term
Relevance
7/24/2026
First Seen
Sources
Codex超高推理模式隐藏成本：一个任务耗尽70%额度的真相
redditr/OpenAI7/12/2026
Related Claims
VerifiedO3 belongs to OpenAI's o-series reasoning models and introduces a Chain-of-Thought mechanism for multi-step internal reasoning.80% similarUnverified推理模型的中间思维链步骤在计费层面同样计入token消耗，OpenAI的o1模型思考token有时高达正常输出的数倍79% similarUnverified推理强度机制的慢思考范式源自 OpenAI o1 系列，Anthropic 随后在 Claude 3.7 及更新模型中跟进实现76% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/608398API
curl https://kongchang.com/api/v1/knowledge/claims/608398MCP
get_claim(id=608398)