微软发射MAI-Image-1号,这是第一个内部人工智能图像生成器,现已在基准中名列前十,并进入联合驾驶和Bing系统。 Microsoft launches MAI-Image-1, its first in-house AI image generator, now top 10 in benchmarks and coming to Copilot and Bing.

微软公司推出了MAI-Imagage-1号,这是第一个用文字生成图像的内部人工智能模型,在LMARENA基准中名列前十。 Microsoft has launched MAI-Image-1, its first in-house AI model for generating images from text, which has ranked in the top 10 on the LMArena benchmark. 该模型是为摄影现实设计,在照明、景观和详细视觉方面优异,同时减少重复性或通用性产出。 Designed for photorealism, the model excels in lighting, landscapes, and detailed visuals while reducing repetitive or generic outputs. 它的运行速度比较快, 并可集成到Copilot和Bing Image Creator. Built with input from creative professionals, it runs faster than larger models and is set to integrate into Copilot and Bing Image Creator. 微软试图减少对外部AI供应商的依赖并扩大其独立的AI能力。 The release marks Microsoft’s push to reduce reliance on external AI providers and expand its independent AI capabilities.