Back    Zoom +    Zoom -
Alibaba's New Text-to-Image Model Qwen-Image Tops Global AI Open Source Community
Recommend
25
Positive
60
Negative
17
Tongyi Qianwen (Qwen), the large language model (LLM) developed by Alibaba Cloud under BABA-W (09988.HK)(BABA.US), open-sourced a new text-to-image model, Qwen-Image, which immediately topped the list of the world's largest AI open source community, HuggingFace, becoming the most prevalent open source model globally.

Qwen-Image is a 20-billion-parameter MMDiT model and is the first foundational image generation model in the Tongyi Qianwen series. It achieved top performance in mainstream evaluation rankings for general image generation such as GenEval and DPG, as well as image editing like GEdit and ImgEdit, with a particular proficiency in rendering Chinese characters. The model is slated to be launched this month.

Related NewsBOCI Lists Actively-traded Stocks for 'Southbound Funds' YTD (Table)

AASTOCKS Financial News
Website: www.aastocks.com