Unverified50% confidenceFactExact time
Zonos 2 采用 MoE 混合专家架构，总参数量达 8B，激活参数仅为 900M
1
Sources
50%
Confidence
Long-term
Relevance
7/11/2026
First Seen
Sources
智谱GLM-5.2全量开放，AI监管风暴同步来袭
bilibili黑鸦Heya6/13/2026
Related Claims
UnverifiedPangu 2.0 Flash adopts a MoE (Mixture of Experts) architecture with a parameter scale of 92B A6B (92B total parameters, 6B activated parameters).64% similarUnverified借助llama.cpp、Ollama等推理框架，普通消费级GPU甚至仅凭CPU即可运行7B至14B参数模型63% similarUnverifiedTypical representatives of the MoE architecture include Mixtral 8x7B and DeepSeek-V2.61% similarUnverifiedKimi K3 uses a MoE (Mixture of Experts) architecture, configured with 896 experts, with 16 activated each time.59% similarVerifiedDeepSeek V4 Flash uses a Mixture of Experts (MOE) architecture with a total parameter count exceeding 600 billion (600B+).59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/486908API
curl https://kongchang.com/api/v1/knowledge/claims/486908MCP
get_claim(id=486908)