答案已发布2周前Last edited 2周前23 来源

Google I/O 2026 重磅发布：Gemini Omni 与 Gemini 3.5 Flash 全面解析

在2026年5月19日的Google I/O大会上，谷歌发布了两个重磅AI模型：Gemini Omni（多模态视频生成模型，可从文本、图片、音频和视频输入生成最长约10秒的视频）和Gemini 3.5 Flash（谷歌首个融合前沿智能与自主行动能力的模型）。 Gemini Omni Flash 已作为Omni系列首个生产级版本于6月底进入公开预览，支持对话式多轮编辑，可在YouTube Shorts内免费使用，在Gemini应用内则需订阅Google AI Plus/Pro/Ultra。

使用 Studio Global AI 搜索并核查事实浏览更多热门页面

AI-generated editorial hero image showing Google Gemini Omni and Gemini 3.5 Flash branding, representing Google's new AI media models announced at Google I/O 2026. — Search & fact-check with cited sources for What new AI media models did Google announce on Tuesday, what are their names, capabilities, speeGoogle announced two new AI media models at Google I/O 2026: Gemini Omni for multimodal video generation and Gemini 3.5 Flash for frontier agentic intelligence.
AI 提示
Create a landscape editorial hero image for this Studio Global article: Search & fact-check with cited sources for What new AI media models did Google announce on Tuesday, what are their names, capabilities, spee. Article summary: On Tuesday, May 19, 2026 (Google I/O), Google announced two major new AI media models: **Gemini Omni** (a multimodal video generation model) and **Gemini 3.5 Flash** (a frontier intelligence + agentic model). A third, **. Topic tags: general, general web, user generated, documentation. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks,
openai.com

在2026年5月19日的Google I/O大会上，谷歌宣布了两款方向截然不同的重磅AI模型：Gemini Omni——一款能从任何输入生成视频的多模态生成模型；以及Gemini 3.5 Flash——谷歌首款将前沿智能与自主行动能力（Agentic）相结合的模型。此外，Omni系列的首个生产级版本Gemini Omni Flash已于6月底进入公开预览。以下是对这些新发布的功能、价格、可用渠道和行业应用的完整梳理。

Gemini Omni — 从任意输入到视频生成

模型概述

Gemini Omni 能够接受文本、图片、音频和现有视频作为输入，生成带有同步音频、画质出色的短视频片段，并基于Gemini的丰富世界知识进行内容生成。首个版本 Gemini Omni Flash 原生输出最长约10秒的视频片段，图片和音频输出功能则计划在未来推出。该模型支持对话式的多轮编辑：你先生成一个片段，然后用自然语言进行修改，系统会在多次修改中保持角色一致性和物理规律。

Studio Global AI

Search, cite, and publish your own answer

Use this topic as a starting point for a fresh source-backed answer, then compare citations before you share it.

使用 Studio Global AI 搜索并核查事实

人们还问