Unverified50% confidenceFactExact time
Qwen1.5的MoE版本采用4个共享专家加60个细粒度专家、每次激活4个的配置
1
Sources
50%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
Partially VerifiedQwen3 Max adopts a MoE (Mixture of Experts) architecture.63% similarUnverifiedQwen3.6-27B采用稠密（Dense）架构，而非MoE（Mixture of Experts）架构61% similarUnverifiedV4-Pro uses a Mixture of Experts (MoE) architecture with a total parameter count of 1.6 trillion (1.6T) and 49 billion (49B) active parameters.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/585430API
curl https://kongchang.com/api/v1/knowledge/claims/585430MCP
get_claim(id=585430)