Unverified50% confidenceFactExact time
Google's Switch Transformer and Mixtral 8x7B are classic representatives of the MoE architecture, and Qwen3, DeepSeek-V3, and others also adopt this architecture.
1
Sources
50%
Confidence
Long-term
Relevance
9/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedGoogle 的 Switch Transformer（2021）和 Mistral 的 Mixtral 8x7B（2023）都采用了稀疏 MoE 架构，即每次前向传播只激活少数几个专家74% similarUnverifiedSwitch Transformer and the Mixtral series have widely adopted the MoE architecture.73% similarVerifiedThe Qwen 3.5 architecture uses an improved Transformer variant that incorporates technologies such as Grouped Query Attention (GQA) and Rotary Position Embedding (RoPE).69% similarUnverifiedTypical representatives of the MoE architecture include Mixtral 8x7B and DeepSeek-V2.65% similarUnverifiedQwen (Tongyi Qianwen) is developed by Alibaba Cloud and is based on the Transformer Decoder architecture.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/848675API
curl https://kongchang.com/api/v1/knowledge/claims/848675MCP
get_claim(id=848675)