Unverified50% confidenceFactExact time
Gemma 4采用混合专家架构（MoE），E2B后缀表示Effective 2B，每次推理只激活约20亿有效参数
1
Sources
50%
Confidence
Long-term
Relevance
7/14/2026
First Seen
Sources
GDScript+Vulkan在Godot中运行Gemma大语言模型实验
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.77% similarUnverifiedGemma 4's E2B version performs comparably to Gemma 3's 27.1-billion-parameter model on multiple tasks in community benchmark tests.77% similarUnverifiedThe Gemma 2B model has approximately 2 billion parameters.72% similarUnverifiedGemma 26B-A4B adopts a Top-K routing strategy, activating only K experts per token (typically K=2).68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/504401API
curl https://kongchang.com/api/v1/knowledge/claims/504401MCP
get_claim(id=504401)