Unverified50% confidenceFactExact time
Low-confidence requests first fail on the small model and then fall back to the large model with the same context for a second generation, causing such requests to incur the cost of two generations and resulting in double payment.
1
Sources
50%
Confidence
Long-term
Relevance
9/8/2026
First Seen
Sources
Related Entities
Related Claims
Unverified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69% similarUnverifiedThe alignment tax refers to the quantifiable cost that alignment measures impose on a model's actual performance, such as decreased accuracy on objective tasks, over-refusal, and homogeneous outputs.67% similarUnverified当Cursor检测到账号使用异常时，可能会将请求路由到性能较低的模型版本或降低响应质量，这被称为模型降智67% similarUnverifiedTwo models with similar intelligence scores may differ in inference cost by several times or even an order of magnitude.66% similarUnverifiedLarge model outputs are probabilistic and non-deterministic; running the same prompt on the same version of a model multiple times may produce different results, making quality judgments in gradual rollouts difficult.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/875660API
curl https://kongchang.com/api/v1/knowledge/claims/875660MCP
get_claim(id=875660)