Unverified50% confidenceFactExact time
The Iris model adopts a Mixture-of-Experts (MoE) architecture, where 35B-A3B indicates a total parameter count of 35 billion and an activated parameter count of 3 billion per inference.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedGemma 4's 26B MoE model uses a Mixture of Experts architecture, with 26 billion total parameters but only approximately 4 billion parameters active at any given moment.78% similarUnverifiedAgent A1 是一个350亿参数的MoE模型，每次推理仅激活约30亿参数75% similarVerifiedHy3 adopts a Mixture of Experts (MoE) architecture with a total parameter count of 295 billion (295B), activating only 21 billion (21B) parameters per inference.73% similarUnverified混元3是混合专家模型（MoE），总参数量为295B，激活参数约21B70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/915222API
curl https://kongchang.com/api/v1/knowledge/claims/915222MCP
get_claim(id=915222)