Unverified50% confidenceFactExact time
Large models are stateless; in multi-turn conversations the client must resend the complete history of messages with each request, causing the input length of the Nth turn to be roughly the accumulation of all messages from the previous N-1 turns, so input cost grows quadratically with the number of turns.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedUnder a linear conversation structure, token consumption is cumulative—every request requires retransmitting the entire history, increasing inference latency and driving up API call costs.76% similarUnverifiedBecause every conversation turn re-reads all prior content, the longer a session runs, the more expensive each new message becomes—costs accumulate with session length rather than instruction length.73% similarUnverifiedA longer context makes model conversations more coherent, but also consumes more tokens and may hit the model's maximum input length limit.72% similarUnverified随着对话轮次增加，每轮需携带完整历史上下文，输入token数量持续膨胀，导致推理时间呈现非线性增长70% similarUnverified多轮对话中第N轮的输入成本是前N-1轮内容的总和，累积效应会导致成本抛物线式增长70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1025621API
curl https://kongchang.com/api/v1/knowledge/claims/1025621MCP
get_claim(id=1025621)