Google DeepMind 推出 Gemini Omni 1.2 Flash，支持生成更长视频

发生了什么

Google DeepMind 正通过 Gemini API 和面向企业的 Gemini 推出 Gemini Omni 1.2 Flash，新增原生视频生成与编辑功能，可处理最长 15 秒的视频片段，而 Omni 1.1 最长为 10 秒。帖子称，其价格比 Omni 1.1 低 30%，生成 720p 视频的费用为每秒 $0.07，并介绍了参考内容遵循度、对话式编辑、语音和一致性的改进。

接下来读什么

• 今日还有哪些重要变化OpenAI 发布 Images 2.6，支持导出 SVG、生成 GIF，速度更快

3 家信源报道

这件事到目前为止

Google is rolling out Gemini Omni 1.2 Flash for video generation and editing through the Gemini API and Gemini for business, raising the clip limit to 15 seconds from 10 seconds in Omni 1.1. Google says 720p video costs $0.07 per second, 30% less than Omni 1.1. Google says the update improves reference adherence, conversational editing, speech and consistency, including how scenes hold together and visual glitches such as jump cuts. Logan Kilpatrick also cited cinematic camera POV controls and less visual drift. In Google Vids, users with a Google account or Google Workspace subscription can generate 15-second clips at no additional cost; video creation in the Gemini app is rolling out to Google AI Ultra subscribers.