Unverified50% confidenceBenchmarkExact time
V4-Flash-Vision-Exp在纯文本Agent能力上与V4-Flash基本持平
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/3/2026
First Seen
Valid until: 12/2/2026
Sources
Related Entities
Related Claims
UnverifiedV4-Flash-Vision-Exp能够读取截图、解析图表，并结合工具调用完成多模态Agent任务79% similarUnverifiedDeepSeek将实验性多模态模型V4-Flash-Vision-EXP上线API平台，其文本能力与V4 Flash持平78% similarUnverified8月3日DeepSeek在HuggingFace上开源了V4系列首款实验性多模态视觉模型DeepSeek-V4-Flash-Vision-Exp74% similarUnverified该模型基于V4 Flash架构，通过持续训练（continued training）叠加视觉模块获得图像理解能力73% similarUnverified据B站UP主九天实测，DeepSeek V4 Flash搭配DSH V0.1.1的原生视觉能力可产出前端页面、HTML版PPT，并一句话生成接近商业级水准的宣传短片67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/849886API
curl https://kongchang.com/api/v1/knowledge/claims/849886MCP
get_claim(id=849886)