Unverified60% confidenceFactTime unknown
Kimi K2 Thinking uses a Mixture of Experts (MoE) architecture with total parameters reaching the trillion scale, activating only a subset of expert networks per inference.
2
Sources
60%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
Verified1.6万亿参数级别的巨型模型通常采用混合专家架构（MoE），每次推理只激活一小部分专家子网络80% similarVerifiedThe MoE architecture uses a gating network to dynamically select the K most relevant experts (typically Top-K, K=2) for each input token, activating only approximately 10%–30% of total parameters during inference.77% similarUnverifiedThe MoE architecture activates only the Top-K experts per inference, so the total parameter count far exceeds the number of parameters actually activated in a single inference.75% similarUnverifiedKimi K3 adopts a Mixture of Experts (MoE) architecture, activating only a subset of parameter sub-networks during inference to reduce computational costs.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24214API
curl https://kongchang.com/api/v1/knowledge/claims/24214MCP
get_claim(id=24214)