Unverified50% confidenceFactExact time
Stable Diffusion 3和Flux等文生图模型采用Flow Matching框架
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/15/2026
First Seen
Valid until: 10/13/2026
Sources
Related Claims
UnverifiedStable Diffusion 3和Meta的MovieGen等大型生成模型已转向基于流匹配的框架80% similarVerifiedFlow Matching通过学习确定性速度场将噪声分布连续映射到数据分布，Stable Diffusion 3和Flux均采用此框架80% similarVerified扩散模型成为Stable Diffusion、DALL-E 3等主流工具的底层架构75% similarUnverifiedDALL·E 2和Stable Diffusion均基于扩散模型原理72% similarUnverified扩散模型的数学本质是基于随机微分方程和得分匹配理论，支撑着Stable Diffusion、DALL-E等图像生成系统72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/519924API
curl https://kongchang.com/api/v1/knowledge/claims/519924MCP
get_claim(id=519924)