模型 profile

Veo AI 视频生成器

Google DeepMind's flagship video generation family for cinematic clips, image-referenced video, vertical output, native audio, and API workflows.

首页/模型s/Veo

是否适合使用 Veo?

适合用来做： cinematic generation, image-referenced clips, vertical short-form video, Google ecosystem users, professional upscaling

Skip or compare first 的情况: Availability varies by country, product surface, and plan. Veo 3.1 is preview in the Gemini API docs and may change. Exact consumer plan limits need recurring checks.

定位： Google DeepMind's flagship video generation model family for cinematic video, native audio, reference images, extension, and Google ecosystem workflows.

核心能力

  • Ingredients to VideoVeo 3.1 lets users create video from reference images with richer movement and control.来源
  • Native vertical outputVeo 3.1 supports native 9:16 generation for short-form platforms.来源
  • 1080p and 4K upscalingThe 3.1 update adds higher-fidelity resolution options for production workflows.来源
  • API accessGoogle documents video generation through the Gemini API and Vertex AI ecosystem.来源

公司摘要

Built by Google DeepMind

Founded / HQ: 2010 / United States / United Kingdom

What it does: Google DeepMind is Google's frontier AI research and product lab, combining the former DeepMind and Google Brain organizations into a single AI team. confirmed

Main products/ecosystem: Veo, Veo 2, Veo 3, Veo 3.1, Flow confirmed

Founders/leadership: partial/evidence pending confirmed

Scale/funding/status: Status: Google Subsidiary confirmed

Relationship to Veo: Google DeepMind develops Veo, while Google distributes it through Gemini, Flow, Google Vids, Gemini API, and Vertex AI-style developer surfaces.

价格、入口与开源状态

访问方式： Closed Web, Closed Api

开源/闭源： Closed. Veo weights are not open.

Gemini API pricing lists Veo 3 at USD 0.40 per second, Veo 3 Fast at USD 0.15 per second, and Veo 2 at USD 0.35 per second; check page for Veo 3.1 pricing before display.

Price, regional access, and commercial terms need a live check before cost-sensitive recommendations.

最新版本

Veo negative prompt support

2026-07-10 / confirmed

Runway API adds optional negativePrompt for veo3, veo3.1, and veo3.1_fast text-to-video and image-to-video requests.

最新消息

Runway API changelog: Veo negative prompt support

2026-07-10 / Runway Dev

Runway API adds optional negativePrompt for veo3, veo3.1, and veo3.1_fast video requests.

打开原文

为什么排在这里

排名： #6 86

分数来源： Google Gemini Omni Flash/Veo-family entries lead the Artificial Analysis audio-video snapshot and rank highly in Arena/Vercel snapshots.. 证据 confidence is high.

排名 评分方法

关键得分驱动

These are the scoring dimensions most responsible for the current ranking position.

  • 模型 Quality Proxy: 95Google Gemini Omni Flash/Veo-family entries lead the Artificial Analysis audio-video snapshot and rank highly in Arena/Vercel snapshots.
  • Capability Depth: 78Supported workflow tags: t2v, i2v, ingredients_to_video, native_audio, upscaling, vertical_video. Depth rewards T2V/I2V/V2V, references, native audio, editing, API, and open weights where p...
  • Version Maturity: 964 version/history entries in the profile; latest public date is 2026-01-13.
  • Web Authority: 94SEO snapshot found product domain gemini.google.com and company domain deepmind.google; sampled sitemap URL count is 315.

替代选择与对比

Use these when price, access, output style, or workflow fit is uncertain.

  • SoraOpenAI's video and audio generation model family, influential for physical realism, controllability, and world-simulation framing.View model对比
  • SeedanceByteDance's frontier video generation family for multimodal, audio-video, short-form, and cinematic creator workflows.View model对比
  • KlingKuaishou's broad AI video and image generation platform for text-to-video, image-to-video, native audio, references, and creator effects.View model对比
  • Runway GenA mature creative video generation platform and model family for production-style image-to-video, text-to-video, editing, and workflow automation.View model对比

它是什么

Primary content is rendered into static HTML.

Veo is Google DeepMind's top AI video model family. It is distributed through Gemini, Flow, Google Vids, YouTube-oriented tools, the Gemini API, and Vertex AI, giving it both consumer and developer routes.

Veo's product direction emphasizes cinematic quality, prompt adherence, image-referenced generation, mobile-first formats, and high-resolution output. Veo 3 introduced native audio capabilities, and Veo 3.1 expanded Ingredients to Video with more expressive image-conditioned generation, vertical output, and upscaling to 1080p and 4K.

For consumer selection, Veo should be compared on quality, availability, price, generation speed, output length, and access friction. Its strongest advantage is integration with the Google ecosystem and professional workflow surfaces.

版本进展

  • 2026-07-10 / Runway Dev / confirmedVeo negative prompt supportRunway API adds optional negativePrompt for veo3, veo3.1, and veo3.1_fast text-to-video and image-to-video requests.
  • 2026-07 / Google DeepMind / confirmedVeo model cardsOfficial DeepMind model-card index for Veo safety/version updates.
  • 2026-04-08 / Google DeepMind / confirmedVeo 3.1 Lite model cardGoogle DeepMind published the Veo 3.1 Lite model card for the text/image-to-video system with audio.
  • 2026-04-02 / Google / confirmedVeo 3.1 in Google Vids free monthly accessGoogle announced that personal Google accounts get monthly no-cost Veo 3.1 video generations in Google Vids, with higher limits for Google AI Pro/Ultra and Workspace AI Ultra accounts.
  • 2026-01-13 / Google / confirmedVeo 3.1Improves Ingredients to Video. Adds native vertical outputs. Adds 1080p and 4K upscaling options.
  • 2026-01 / current_confirmed / confirmedVeo 3.1 Preview / Fast PreviewGemini API model codes include veo-3.1-generate-preview and veo-3.1-fast-generate-preview. Supports text, image, last-frame, reference-image, and video-extension inputs depending on model.
  • 2025-07-17 / historical / confirmedVeo 3 PreviewGemini API changelog lists video with audio generation for Veo 3 preview.
  • 2024-05 / historical / confirmedVeoVeo announced at Google I/O 2024 as Google DeepMind's generative video model.

开源 / 来源证据

Status: Closed

Veo weights are not open.

最新消息

模型-specific news entries link directly to the original publisher; article bodies stay off-site.

完整得分拆解

权重ed evidence dimensions separate model quality from access, web authority, ecosystem, and source confidence. See methodology and rankings.

维度Score权重证据Confidence
模型 Quality Proxy 95 24% Google Gemini Omni Flash/Veo-family entries lead the Artificial Analysis audio-video snapshot and rank highly in Arena/Vercel snapshots. high
Missing: live Artificial Analysis rank/Elo by modality, live Arena AI rank/score by modality
Capability Depth 78 16% Supported workflow tags: t2v, i2v, ingredients_to_video, native_audio, upscaling, vertical_video. Depth rewards T2V/I2V/V2V, references, native audio, editing, API, and open weights where present. high
Missing: hands-on feature verification, mode-specific limits by duration/resolution
Access And Pricing 78 12% Access types: closed_web, closed_api. Pricing note: Consumer access depends on Gemini/Flow/Workspace plans. API pricing depends on Gemini API or Vertex AI model, output duration, and resolution. high
Missing: current free tier, normalized price per video minute
Version Maturity 96 12% 4 version/history entries in the profile; latest public date is 2026-01-13. high
Missing: automated release-note monitor, model ID/version mapping across providers
Ecosystem Popularity 68 10% GitHub stars are not applicable to this closed or research-only product unless an official public model repository is confirmed.

No public source link is attached yet.

 medium
Missing: GitHub: n/a for closed-source model
Web Authority 94 10% SEO snapshot found product domain gemini.google.com and company domain deepmind.google; sampled sitemap URL count is 315. medium
Missing: Similarweb/API traffic, Tranco rank
公司 Distribution 80 10% Google DeepMind distribution profile: Veo is a flagship option for consumers who want cinematic realism, strong prompt understanding, Google product integration, and increasingly native audio/video capabilities through... medium
Missing: app store ratings/review volume, product MAU/traffic
来源 Confidence 100 6% Profile source confidence is confirmed with 3 official URL(s), 4 feature evidence item(s), SEO present, GitHub n/a. high
Missing: paid/API evidence refresh, manual output review artifacts

限制与价格

Consumer access depends on Gemini/Flow/Workspace plans. API pricing depends on Gemini API or Vertex AI model, output duration, and resolution.

  • Availability Varies By Country, Product Surface, And Plan.
  • Veo 3.1 Is Preview In The Gemini API Docs And May Change.
  • Exact Consumer Plan Limits Need Recurring Checks.

来源