Unverified50% confidenceFactExact time
The token cost of LLM collectives grows approximately linearly or even superlinearly with the number of nodes and interaction rounds.
1
Sources
50%
Confidence
Long-term
Relevance
10/20/2026
First Seen
Sources
Related Entities
Related Claims
Unverified推理成本（Inference Cost）是大模型持续失血的主要原因，随用户规模呈线性甚至超线性增长67% similarUnverified在反向模式下，当模型参数数量高达数百万时，计算梯度的成本仅与一次前向计算相当66% similarUnverified推理成本随着用户规模增长而线性甚至超线性上升，使得即使收入高速增长，AI公司仍可能持续大额亏损。64% similarUnverified在百万级样本的批量预测场景中，LLM的推理成本可能比树模型高出三到四个数量级62% similarUnverifiedImprovements in intelligence are often accompanied by nonlinear cost increases; upgrading from a mid-tier model to a top-tier model may raise benchmark scores by only a few percentage points, but calling costs can multiply several times over.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1073935API
curl https://kongchang.com/api/v1/knowledge/claims/1073935MCP
get_claim(id=1073935)