Unverified60% confidenceFactExact time
The model has 512 experts per layer, but only 10 are activated per token, meaning the weights actually participating in computation are approximately 6B.
2
Sources
60%
Confidence
Long-term
Relevance
10/5/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedMoE的8选2配置意味着8个专家中每次只激活2个，实际计算量仅为密集模型的约四分之一74% similarUnverifiedMoE models activate only approximately 10 out of 512 experts per token, with an effective activated parameter count of approximately 20–30B.72% similarUnverifiedM2.5 adopts a Mixture of Experts (MoE) architecture, where only 10B parameters are actually involved in computation during each inference, while the total parameter count may far exceed 10B.68% similarUnverified服务同等数量的并发用户，超大模型所需的GPU集群规模可能是同等智能水平小模型的5至10倍67% similarUnverifiedTaking a 1000×1000 weight matrix as an example, if the LoRA rank is set to 8, only approximately 16,000 parameters need to be trained instead of the original 1,000,000 parameters.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/978653API
curl https://kongchang.com/api/v1/knowledge/claims/978653MCP
get_claim(id=978653)