Unverified95% confidenceFactTime unknown
Meta开源了SAM3（Segment Anything Model 3），支持通过自然语言文字描述进行视觉分割
1
Sources
95%
Confidence
Medium-term (~90 days)
Relevance
5/31/2026
First Seen
Valid until: 8/29/2026
Sources
SAM3深度解析：Meta开源视觉分割模型如何用自然语言理解现实世界
bilibili碳基玩家Bryce
Related Entities
Related Claims
UnverifiedSAM 3实现了Meta所称的可提示概念分割（Promptable Concept Segmentation），只需给出一个简短的名词短语即可分割图像中该概念的所有实例74% similarUnverifiedSAM 3引入文本驱动的概念分割，消除了对空间提示的依赖66% similarUnverifiedMeta推出了Llama 3.2 1B/3B小语言模型63% similarUnverifiedMeta's Chameleon and ByteDance's SEED-X have explored unified vision-language frameworks using discrete tokenization schemes63% similarUnverifiedAnthropic的三档模型命名Opus、Sonnet、Haiku源自音乐和文学术语59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/10204API
curl https://kongchang.com/api/v1/knowledge/claims/10204MCP
get_claim(id=10204)