China's SenseTime to Launch Image Generation Model U1 Pro, Rivaling GPT-Image 2

en.Wedoany.com Reported - SenseTime is secretly developing a multimodal large model named "U1 Pro," led by co-founder and chief scientist Lin Dahua. Targeting the "design" scenario, the model aims to rival OpenAI's GPT-Image 2 as an image generation model with reasoning capabilities.

Developed by SenseTime Research Institute as part of the SenseNova model family, the model is expected to launch internal invite-only testing and provide services to clients by July this year. According to insiders, "U1 Pro" can handle complex design requirements like a "thinking designer," enabling a long-cycle process of design, generation, and review, while supporting 8K resolution output. In extensive internal evaluations, images generated by "U1 Pro" under the same prompts closely matched or even surpassed those from GPT-Image 2 in quality.

In the text-to-image scoring of the LMSYS Chatbot Arena, GPT-Image 2 significantly outperformed Google's Nano Banana 2 in image quality, text rendering, and instruction following, sparking widespread acclaim in the design industry. Industry insiders generally predict that OpenAI will soon release a new AI image generation model, with "design" as a key focus area.

The leak of SenseTime's new "U1" model signals to the outside world that beyond programming, led by top AI companies like Anthropic and Zhipu AI, "design" is becoming the next major arena for multimodal models.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com