Unverified50% confidenceFactExact time
Due to the cyclic reasoning nature of Agent systems, the complete conversation history is resent to the model after each tool call, causing Token consumption to grow super-linearly.
1
Sources
50%
Confidence
Long-term
Relevance
9/9/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEach model response includes the full conversation history (context window) in billing, and as the number of dialogue turns increases, the token consumption per call grows linearly or even super-linearly.80% similarUnverifiedAfter decomposition and planning, an Agent may trigger dozens of LLM inference calls, with cumulative token consumption reaching tens of times that of a direct question-and-answer scenario.77% similarUnverifiedIn each iteration of the loop, an agent needs to repackage the system prompt, conversation history, tool descriptions, and all outputs from previous steps before feeding them into the model, causing token consumption to grow linearly or even super-linearly with the number of steps.76% similarUnverifiedIn multi-turn conversations, before each new response is generated, the system must re-input all prior conversation history into the model, causing token consumption to accumulate and grow with each turn76% similarUnverifiedThe cost of each large language model call is positively correlated with the number of input tokens, and in an Agent loop, as the conversation history accumulates, the cost of each call grows linearly.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/882582API
curl https://kongchang.com/api/v1/knowledge/claims/882582MCP
get_claim(id=882582)