Unverified50% confidenceFactExact time
在Agent应用中，每一轮推理-行动-观察循环都需要将完整历史上下文重新输入模型，导致Token消耗随任务复杂度呈非线性增长
1
Sources
50%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
Claude Sonnet 5发布：最强Agent能力的中端AI模型
twitterclaudeai6/30/2026
Related Claims
UnverifiedMost Agent frameworks adopt a full context replay strategy, requiring the entire conversation history to be passed in with each call, causing token consumption to grow quadratically with the number of loop iterations.80% similarUnverified典型的Agent框架如ReAct、AutoGPT、LangGraph采用感知-推理-行动的迭代模式，上下文随轮次累积导致后期每轮Token消耗呈线性甚至指数增长78% similarUnverifiedWhen an agent is in a loop, each call carries the accumulated conversation history, causing token consumption to grow linearly or even super-linearly with the number of iterations.77% similarUnverifiedAgent 的每一轮循环都需将完整历史上下文传入 LLM，导致 token 消耗随任务复杂度呈线性乃至指数增长75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/129566API
curl https://kongchang.com/api/v1/knowledge/claims/129566MCP
get_claim(id=129566)