This start-up claims its 3D AI model's better than Google
What's the story
Beijing-based artificial intelligence (AI) start-up Vast is challenging tech giants like Tencent Holdings and Google in the race to develop 3D AI models. These models permit creators to generate three-dimensional visual content from text or image inputs. The company was founded in 2023 by Simon Song Yachen, a co-founder of Chinese AI unicorn MiniMax.
Vision
Vast's mission and latest model
Vast's mission is to enable everyone to create 3D content without any barriers or costs. The company also plans to develop a TikTok-like platform for user-generated 3D content. With more than 100 employees in Beijing and Hangzhou, Vast operates an AI 3D generation platform called Tripo Studio. Last week, they launched the latest iteration of this platform, the Tripo 3.0 model which promises "the most accurate, detail-rich geometry yet."
Industry standing
Claiming to be global leader in AI 3D large models
Song, who previously worked at Chinese AI pioneer SenseTime, claims that Vast is a global leader in the field of AI 3D large models. He founded the company shortly after Google launched its text-to-3D model DreamFusion in 2022. This development opened up new possibilities for generating 3D content, and has since become a major focus area for companies looking to attract professional creators from gaming and film industries as well as individual artists.
Market competition
Tencent's advancements in AI-generated 3D content
In March, Chinese social media as well as gaming giant Tencent made five of its early Hunyuan 3D models open-source. Last month, it launched and open-sourced its Hunyuan 3D world model 1.0, capable of generating a 3D virtual world from a line of text or an image in minutes. This process would have taken weeks for a professional modeling team to complete earlier.