Unverified55% confidenceFactTime unknown
MiniMind-V的整体架构设计参考了LLaVA（Large Language and Vision Assistant）的经典范式
1
Sources
55%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDoubao 2.0 Mini supports Vision Understanding functionality, with the underlying layer using a Vision Transformer (ViT) architecture to encode images into token sequences.73% similarUnverifiedTongying Lab released a Vision-Language-Action (VLA) model.66% similarUnverifiedThe core strengths of the MiniMax M3 model are its ultra-long context processing capability and multimodal comprehension capability (capable of handling text, images, and video).65% similarUnverified蓝图是虚幻引擎的可视化脚本系统，底层编译为字节码由UE虚拟机解释执行，性能略低于原生C++但开发效率更高65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/29984API
curl https://kongchang.com/api/v1/knowledge/claims/29984MCP
get_claim(id=29984)