Verified65% confidenceFactTime unknown
GLM-4.6采用专家混合架构（MoE），总参数量达到3550亿，推理时只激活约320亿个参数
3
Sources
65%
Confidence
Medium-term (~90 days)
Relevance
6/1/2026
First Seen
Valid until: 8/30/2026
Sources
GLM-4.6深度实测：性能、价格与编程能力全面评估
bilibili薛定猫AI
Related Entities
Related Claims
UnverifiedGLM-4.7是一款MOE（混合专家）架构模型，总参数量达358B（3580亿）86% similarUnverifiedHy3采用混合专家（MoE）架构，总参数量达2950亿（295B），每次推理仅激活210亿（21B）参数78% similarUnverifiedLFM2.5 has 8.3 billion total parameters but only 1.5 billion parameters activated per inference75% similarVerifiedGemma 4的26B MOE模型采用混合专家架构，拥有260亿总参数但任意时刻仅约40亿参数处于激活状态75% similarUnverified智谱GLM系列从GLM4.5到4.7在358B参数规模上连续打磨三代，随后GLM5翻倍至754B74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/24184API
curl https://kongchang.com/api/v1/knowledge/claims/24184MCP
get_claim(id=24184)