Unverified50% confidenceFactExact time
Active parameters describe the portion involved in computation per token, not the model's download size or the memory required to load it.
1
Sources
50%
Confidence
Long-term
Relevance
9/29/2026
First Seen
Sources
Related Entities
Related Claims
Unverified提示缓存缓存的是模型处理输入时产生的中间计算结果（KV对），而非模型的输出72% similarUnverified该模型的架构设计目标是让每个token的解码速度只取决于激活参数而非总参数72% similarUnverifiedTotal parameter count reflects the upper limit of a model's knowledge capacity and expressive power, while activated parameter count determines the computational cost and latency during inference.72% similarVerifiedSpeculative decoding is a lossless acceleration technique that does not alter the model's output distribution; it uses a draft model with far fewer parameters than the target model to generate candidate tokens, which the target model then verifies in parallel.72% similarUnverified部署MOE模型时必须按总参数量而非激活参数量来估算显存需求，因为所有参数都需要在高速存储中随时可访问70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/958356API
curl https://kongchang.com/api/v1/knowledge/claims/958356MCP
get_claim(id=958356)