Unverified50% confidenceFactExact time
据社区讨论，DeepSeek V4将带来原生MXFP4混合专家（MoE）架构，配合超大上下文能力
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/17/2026
First Seen
Valid until: 10/15/2026
Sources
Related Claims
UnverifiedDeepSeek-V4 Flash采用混合注意力、mHC混合机制以及哈希路由的MoE等架构设计69% similarUnverifiedGLM5在部分技术架构上完全复现了DeepSeek V3的设计，采用MoE（混合专家）架构68% similarUnverifiedDeepSeek-Reasonix提供两个基础模型选项：V3 Flash和V3 Pro，并配有五个推理强度级别66% similarVerifiedDeepSeek V4全面适配国产华为昇腾系列AI芯片，摆脱对英伟达GPU的依赖65% similarUnverifiedGemma 4引入了多模态能力，Qwen3支持超长上下文，DeepSeek采用MoE（混合专家）架构65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/544939API
curl https://kongchang.com/api/v1/knowledge/claims/544939MCP
get_claim(id=544939)