**Stability AI 发布重大突破：Stable Diffusion 3 文生图模型引领技术新高度**
伦敦——今日，人工智能研发公司 Stability AI 宣布发布了其最新的 Stable Diffusion 3 研究论文，揭示了这一先进文生图模型的底层技术创新。据论文所述，Stable Diffusion 3 在文本到图像的生成领域取得了显著的提升，尤其在排版和对提示的精准遵循方面，超越了目前市面上的顶级系统，包括 DALL·E 3、Midjourney v6 和 Ideogram v1。
Stable Diffusion 3 的核心在于其创新的多模态扩散Transformer（MMDiT）架构，这一架构为图像和语言表示各自分配了独立的权重集，从而增强了模型在理解文本和拼写准确性上的能力。这一突破性进展意味着用户可以更精确地通过自然语言指令引导模型生成符合预期的高质量图像。
据 Stability AI 的研究显示，Stable Diffusion 3 的性能提升基于人类偏好评估，其生成的图像在与提示信息的匹配度上得到了更高的评价。这一进步不仅对艺术创作、设计和可视化领域产生深远影响，也为未来人工智能在理解和生成复杂视觉内容方面开辟了新的可能。
Stability AI 的这一创新成果再次证明了公司在人工智能领域的领导地位，预示着文生图技术将进入一个全新的发展阶段。随着 Stable Diffusion 3 的发布，业界期待这一技术能为全球用户带来更为智能、精准的图像生成体验，同时也将推动相关行业的创新与进步。
英语如下：
**News Title:** “Stability AI Launches Stable Diffusion 3: Revolutionizing Text-to-Image Technology, Outperforming Advanced Systems like DALL·E 3”
**Keywords:** Stability AI, Stable Diffusion 3, Text-to-image model
**News Content:**
**Stability AI Unveils Groundbreaking Stable Diffusion 3 Text-to-Image Model**
London – Today, AI research firm Stability AI announced the release of its latest Stable Diffusion 3 research paper, detailing the advancements in this cutting-edge text-to-image model. According to the paper, Stable Diffusion 3 has made significant strides in text-to-image generation, surpassing top systems like DALL·E 3, Midjourney v6, and Ideogram v1, particularly in layout and precise adherence to prompts.
At the core of Stable Diffusion 3 lies its innovative Multi-modal Diffusion Transformer (MMDiT) architecture, which allocates separate weight sets for image and language representations, enhancing the model’s ability to understand text and maintain spelling accuracy. This breakthrough advancement enables users to guide the model more precisely with natural language instructions to generate high-quality images that align with expectations.
Stability AI’s research indicates that Stable Diffusion 3’s performance improvements are based on human preference assessments, with its generated images receiving higher ratings for their alignment with prompt information. This progress has far-reaching implications for artistic creation, design, and visualization, as well as opening up new possibilities for AI in understanding and generating complex visual content.
Stability AI’s innovation reasserts the company’s leadership position in the AI sector and signals a new era for text-to-image technology. With the launch of Stable Diffusion 3, the industry anticipates a more intelligent and accurate image generation experience for global users, driving innovation and progress in related industries.
【来源】https://stability.ai/news/stable-diffusion-3-research-paper
Views: 1