Unverified50% confidenceFactExact time
From low to the highest level, the cost and time of effort levels rise approximately exponentially, but the improvement in capability is not proportional.
1
Sources
50%
Confidence
Long-term
Relevance
10/14/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedImprovements 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.72% similarUnverified自2024年以来，多项研究和产业实践显示单纯增加参数量带来的收益正在递减，训练成本呈指数增长而基准测试分数提升趋于平缓70% similarUnverifiedThe marginal returns from moving from the High to Max inference tier are clearly diminishing: the DeepSWE score increases by only approximately 4 percentage points, while the cost increases by approximately 1.5x.67% similarUnverifiedJevons' Paradox states that when resource use efficiency improves and unit costs decline, total consumption actually increases, and is used to explain how token price reductions drive multiplicative growth in AI usage volume.66% similarUnverifiedThe study concludes that for the models and tasks tested, loading the full occupational profile by default neither improves accuracy nor reduces cost—in fact, it significantly raises costs.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1029741API
curl https://kongchang.com/api/v1/knowledge/claims/1029741MCP
get_claim(id=1029741)