xAI 发布 Aurora 图像生成模型，增强 Grok 多模态能力

AI 摘要

xAI 宣布推出代号为 Aurora 的新型自回归图像生成模型，该模型采用混合专家网络，能够处理文本和图像交错数据，支持多模态输入，擅长生成逼真图像和精确遵循文本指令。Aurora 已在 X 平台的部分国家上线，并将在一周内向所有用户开放。xAI 还推出了 Grok Imagine API，提供视频生成、编辑和图像创建功能。

正文节选

We are updating Grok's capabilities with a new autoregressive image generation model, code-named Aurora, available on the 𝕏 platform. A newer model is available Grok Imagine API is here Our most powerful generative model yet — state-of-the-art video generation, video editing, and image creation, all in one API. We've enhanced Grok's image generation abilities with a new model, code-named Aurora. Aurora is an autoregressive mixture-of-experts network trained to predict the next token from interle

发布时间：—

抓取时间：2026-08-02 17:33

来源机构：xAI