多模型层

用一个 API 接入全球顶级视频、图像、音乐与 LLM 模型

同样的模型，更优的价格。比官方 API 更便宜。快速，简单。

探索 AI API API 文档

受到信赖：120,000+ 开发者 · 80+ 国家/地区

99.9%

正常运行时间

24.9s

响应时间

24/7

支持

#1

数据安全

平台实时状态获取支持→

透明定价，为规模化而生

以更低成本接入领先 AI 模型。一个透明钱包，无隐藏费用，失败生成不收费。

比较所有模型

平均节省

最高 84%

相比官方 API

失败任务

$0

零风险计费政策

隐藏费用

无

一个透明的钱包

Model

Category

KIE — vs — Official

Google Nano Banana 2 4K

Image

KIE$0.09per image

−43.8%vs

Official$0.16per image

Google Nano Banana 2 2K

Image

KIE$0.06per image

−50.0%vs

Official$0.12per image

Google Nano Banana 2 1K

Image

KIE$0.04per image

−50.0%vs

Official$0.08per image

Google Nano Banana Pro 1/2K

Image

KIE$0.09per image

−40.0%vs

Official$0.15per image

Google Nano Banana Pro 4K

Image

KIE$0.12per image

−60.0%vs

Official$0.30per image

GPT Image 2, Image-to-Image, 4K

Image

KIE$0.08per image

−80.6%vs

Official$0.413per image

GPT Image 2, Image-to-Image, 2K

Image

KIE$0.05per image

−78.6%vs

Official$0.234per image

GPT Image 2, Image-to-Image, 1K

Image

KIE$0.03per image

−86.3%vs

Official$0.219per image

GPT Image 2, Text-to-Image, 4K

Image

KIE$0.08per image

−80.6%vs

Official$0.413per image

GPT Image 2, Text-to-Image, 2K

Image

KIE$0.05per image

−78.6%vs

Official$0.234per image

GPT Image 2, Text-to-Image, 1K

Image

KIE$0.03per image

−86.3%vs

Official$0.219per image

Google Veo 3.1 Text-to-Video Quality 1080p

Video

KIE$1.28per video

−60.9%vs

Official$3.20per video

Google Veo 3.1 Image-to-Video Quality 1080p

Video

KIE$1.28per video

−60.9%vs

Official$3.20per video

Google Veo 3.1 Text-to-Video Quality 4K

Video

KIE$1.85per video

−61.4%vs

Official$4.80per video

Google Veo 3.1 Image-to-Video Quality 4K

Video

KIE$1.85per video

−61.4%vs

Official$4.80per video

Grok Imagine Image-to-Video 720p

Video

KIE$0.015per sec

−78.6%vs

Official$0.07per sec

Grok Imagine Text-to-Video 720p

Video

KIE$0.015per sec

−78.6%vs

Official$0.07per sec

Grok Imagine Image-to-Video 480p

Video

KIE$0.008per sec

−84.0%vs

Official$0.05per sec

Grok Imagine Text-to-Video 480p

Video

KIE$0.008per sec

−84.0%vs

Official$0.05per sec

Claude Opus 4.7 Input

LLM

KIE$1.425per 1M tokens

−71.5%vs

Official$5.00per 1M tokens

Claude Opus 4.7 Output

LLM

KIE$7.150per 1M tokens

−71.4%vs

Official$25.00per 1M tokens

Claude Sonnet 4.6 Input

LLM

KIE$0.850per 1M tokens

−71.7%vs

Official$3.00per 1M tokens

Claude Sonnet 4.6 Output

LLM

KIE$4.275per 1M tokens

−71.5%vs

Official$15.00per 1M tokens

GPT-5.5 Chat Input

LLM

KIE$1.40per 1M tokens

−72.0%vs

Official$5.00per 1M tokens

GPT-5.5 Chat Output

LLM

KIE$8.40per 1M tokens

−72.0%vs

Official$30.00per 1M tokens

Google Nano Banana 2 4KImage

KIE$0.09per image

−43.8%vs Official

Official$0.16per image

Google Nano Banana 2 2KImage

KIE$0.06per image

−50.0%vs Official

Official$0.12per image

Google Nano Banana 2 1KImage

KIE$0.04per image

−50.0%vs Official

Official$0.08per image

Google Nano Banana Pro 1/2KImage

KIE$0.09per image

−40.0%vs Official

Official$0.15per image

Google Nano Banana Pro 4KImage

KIE$0.12per image

−60.0%vs Official

Official$0.30per image

GPT Image 2, Image-to-Image, 4KImage

KIE$0.08per image

−80.6%vs Official

Official$0.413per image

GPT Image 2, Image-to-Image, 2KImage

KIE$0.05per image

−78.6%vs Official

Official$0.234per image

GPT Image 2, Image-to-Image, 1KImage

KIE$0.03per image

−86.3%vs Official

Official$0.219per image

GPT Image 2, Text-to-Image, 4KImage

KIE$0.08per image

−80.6%vs Official

Official$0.413per image

GPT Image 2, Text-to-Image, 2KImage

KIE$0.05per image

−78.6%vs Official

Official$0.234per image

GPT Image 2, Text-to-Image, 1KImage

KIE$0.03per image

−86.3%vs Official

Official$0.219per image

Google Veo 3.1 Text-to-Video Quality 1080pVideo

KIE$1.28per video

−60.9%vs Official

Official$3.20per video

Google Veo 3.1 Image-to-Video Quality 1080pVideo

KIE$1.28per video

−60.9%vs Official

Official$3.20per video

Google Veo 3.1 Text-to-Video Quality 4KVideo

KIE$1.85per video

−61.4%vs Official

Official$4.80per video

Google Veo 3.1 Image-to-Video Quality 4KVideo

KIE$1.85per video

−61.4%vs Official

Official$4.80per video

Grok Imagine Image-to-Video 720pVideo

KIE$0.015per sec

−78.6%vs Official

Official$0.07per sec

Grok Imagine Text-to-Video 720pVideo

KIE$0.015per sec

−78.6%vs Official

Official$0.07per sec

Grok Imagine Image-to-Video 480pVideo

KIE$0.008per sec

−84.0%vs Official

Official$0.05per sec

Grok Imagine Text-to-Video 480pVideo

KIE$0.008per sec

−84.0%vs Official

Official$0.05per sec

Claude Opus 4.7 InputLLM

KIE$1.425per 1M tokens

−71.5%vs Official

Official$5.00per 1M tokens

Claude Opus 4.7 OutputLLM

KIE$7.150per 1M tokens

−71.4%vs Official

Official$25.00per 1M tokens

Claude Sonnet 4.6 InputLLM

KIE$0.850per 1M tokens

−71.7%vs Official

Official$3.00per 1M tokens

Claude Sonnet 4.6 OutputLLM

KIE$4.275per 1M tokens

−71.5%vs Official

Official$15.00per 1M tokens

GPT-5.5 Chat InputLLM

KIE$1.40per 1M tokens

−72.0%vs Official

Official$5.00per 1M tokens

GPT-5.5 Chat OutputLLM

KIE$8.40per 1M tokens

−72.0%vs Official

Official$30.00per 1M tokens

KIE 精选模型

直接通过我们的 API 使用最受欢迎、表现力最强的一批模型。

VID

Seedance 2

$0.057 / s

Seedance 2.0 on KIE is a multimodal Al video model by ByteDance, optimized for fast and realistic video generation. It supports high-quality virtual human video creation with strong multi-shot consistency, enabling more lifelike and cinematic outputs across scenes.

Online试用模型

Veo 3.1

VID

Veo 3.1

$0.025 / req

Google DeepMind’s upgraded AI video model for realistic motion generation, extended clip duration, multi-image reference control, and synchronized audio output in native 1080p.

Online试用模型

VID

Kling 3.0

$0.07 / s

Kling 3.0 is Kling AI’s video generation model that creates videos from text and images, supports multi-shot storytelling, and produces native audio with cinematic control up to 15 seconds.

Online试用模型

GPT Image 2

IMG

GPT Image 2

$0.03 / img

GPT Image 2 is OpenAI’s next-gen image model built for stronger photorealism, cleaner image editing, sharper text rendering, and more polished product photography.

Online试用模型

Nano Banana 2

IMG

Nano Banana 2

$0.04 / img

Meet Nano Banana 2, Google’s Gemini 3.1 Flash Image model, now available via Kie AI API. Built for developers, it combines lightning-fast speed with Pro-level quality, accurate text rendering, strong character consistency, and scalable image generation and editing workflows.

Online试用模型

VID

Seedance 2

$0.057 / s

Seedance 2.0 on KIE is a multimodal Al video model by ByteDance, optimized for fast and realistic video generation. It supports high-quality virtual human video creation with strong multi-shot consistency, enabling more lifelike and cinematic outputs across scenes.

Online试用模型

Veo 3.1

VID

Veo 3.1

$0.025 / req

Google DeepMind’s upgraded AI video model for realistic motion generation, extended clip duration, multi-image reference control, and synchronized audio output in native 1080p.

Online试用模型

VID

Kling 3.0

$0.07 / s

Kling 3.0 is Kling AI’s video generation model that creates videos from text and images, supports multi-shot storytelling, and produces native audio with cinematic control up to 15 seconds.

Online试用模型

GPT Image 2

IMG

GPT Image 2

$0.03 / img

GPT Image 2 is OpenAI’s next-gen image model built for stronger photorealism, cleaner image editing, sharper text rendering, and more polished product photography.

Online试用模型

Nano Banana 2

IMG

Nano Banana 2

$0.04 / img

Meet Nano Banana 2, Google’s Gemini 3.1 Flash Image model, now available via Kie AI API. Built for developers, it combines lightning-fast speed with Pro-level quality, accurate text rendering, strong character consistency, and scalable image generation and editing workflows.

Online试用模型

VID

Seedance 2

$0.057 / s

Seedance 2.0 on KIE is a multimodal Al video model by ByteDance, optimized for fast and realistic video generation. It supports high-quality virtual human video creation with strong multi-shot consistency, enabling more lifelike and cinematic outputs across scenes.

Online试用模型

Veo 3.1

VID

Veo 3.1

$0.025 / req

Google DeepMind’s upgraded AI video model for realistic motion generation, extended clip duration, multi-image reference control, and synchronized audio output in native 1080p.

Online试用模型

VID

Kling 3.0

$0.07 / s

Kling 3.0 is Kling AI’s video generation model that creates videos from text and images, supports multi-shot storytelling, and produces native audio with cinematic control up to 15 seconds.

Online试用模型

GPT Image 2

IMG

GPT Image 2

$0.03 / img

GPT Image 2 is OpenAI’s next-gen image model built for stronger photorealism, cleaner image editing, sharper text rendering, and more polished product photography.

Online试用模型

Nano Banana 2

IMG

Nano Banana 2

$0.04 / img

Meet Nano Banana 2, Google’s Gemini 3.1 Flash Image model, now available via Kie AI API. Built for developers, it combines lightning-fast speed with Pro-level quality, accurate text rendering, strong character consistency, and scalable image generation and editing workflows.

Online试用模型

VID

Seedance 2

$0.057 / s

Seedance 2.0 on KIE is a multimodal Al video model by ByteDance, optimized for fast and realistic video generation. It supports high-quality virtual human video creation with strong multi-shot consistency, enabling more lifelike and cinematic outputs across scenes.

Online试用模型

Veo 3.1

VID

Veo 3.1

$0.025 / req

Google DeepMind’s upgraded AI video model for realistic motion generation, extended clip duration, multi-image reference control, and synchronized audio output in native 1080p.

Online试用模型

VID

Kling 3.0

$0.07 / s

Kling 3.0 is Kling AI’s video generation model that creates videos from text and images, supports multi-shot storytelling, and produces native audio with cinematic control up to 15 seconds.

Online试用模型

GPT Image 2

IMG

GPT Image 2

$0.03 / img

GPT Image 2 is OpenAI’s next-gen image model built for stronger photorealism, cleaner image editing, sharper text rendering, and more polished product photography.

Online试用模型

Nano Banana 2

IMG

Nano Banana 2

$0.04 / img

Meet Nano Banana 2, Google’s Gemini 3.1 Flash Image model, now available via Kie AI API. Built for developers, it combines lightning-fast speed with Pro-level quality, accurate text rendering, strong character consistency, and scalable image generation and editing workflows.

Online试用模型

Suno v5.5

AUD

Suno v5.5

$0.002 / req

Kie AI Music API is an AI music-generation model that converts text prompts into full vocal and instrumental tracks with natural dynamics and coherent musical progression, and it currently supports the latest V5.5 model for enhanced realism and scalable performance.

Online试用模型

Eleven labs

AUD

Eleven labs

$0.07 / 1K

ElevenLabs Eleven V3 enables expressive multilingual Text to Dialogue with audio tag control, multi-speaker support, and natural delivery, designed for dialogue-driven applications, storytelling tools, and immersive voice experiences.

Online试用模型

Claude opus 4.7

LLM

Claude opus 4.7

$1.425 / M

Claude Opus 4.7 delivers a quantum leap in logical reasoning and creative synthesis. With its next-gen architecture and massive 2-million-token context, it serves as the ultimate high-performance partner for those pushing the boundaries of innovation and discovery.

Online试用模型

Gemini 3.1 Pro

LLM

Gemini 3.1 Pro

$0.5 / M

Gemini 3.1 Pro API is the latest general-purpose LLM developed by Google DeepMind, designed to bridge the gap between high-speed execution and deep logic. It empowers developers to build sophisticated agents with state-of-the-art accuracy in coding, creative writing, and cross-modal analysis.

Online试用模型

GPT 5.5

LLM

GPT 5.5

$0.14 / M

GPT-5.5 is OpenAI’s advanced reasoning model for agentic coding, knowledge work, scientific research, and complex multi-step task execution.

Online试用模型

Suno v5.5

AUD

Suno v5.5

$0.002 / req

Kie AI Music API is an AI music-generation model that converts text prompts into full vocal and instrumental tracks with natural dynamics and coherent musical progression, and it currently supports the latest V5.5 model for enhanced realism and scalable performance.

Online试用模型

Eleven labs

AUD

Eleven labs

$0.07 / 1K

ElevenLabs Eleven V3 enables expressive multilingual Text to Dialogue with audio tag control, multi-speaker support, and natural delivery, designed for dialogue-driven applications, storytelling tools, and immersive voice experiences.

Online试用模型

Claude opus 4.7

LLM

Claude opus 4.7

$1.425 / M

Claude Opus 4.7 delivers a quantum leap in logical reasoning and creative synthesis. With its next-gen architecture and massive 2-million-token context, it serves as the ultimate high-performance partner for those pushing the boundaries of innovation and discovery.

Online试用模型

Gemini 3.1 Pro

LLM

Gemini 3.1 Pro

$0.5 / M

Gemini 3.1 Pro API is the latest general-purpose LLM developed by Google DeepMind, designed to bridge the gap between high-speed execution and deep logic. It empowers developers to build sophisticated agents with state-of-the-art accuracy in coding, creative writing, and cross-modal analysis.

Online试用模型

GPT 5.5

LLM

GPT 5.5

$0.14 / M

GPT-5.5 is OpenAI’s advanced reasoning model for agentic coding, knowledge work, scientific research, and complex multi-step task execution.

Online试用模型

Suno v5.5

AUD

Suno v5.5

$0.002 / req

Kie AI Music API is an AI music-generation model that converts text prompts into full vocal and instrumental tracks with natural dynamics and coherent musical progression, and it currently supports the latest V5.5 model for enhanced realism and scalable performance.

Online试用模型

Eleven labs

AUD

Eleven labs

$0.07 / 1K

ElevenLabs Eleven V3 enables expressive multilingual Text to Dialogue with audio tag control, multi-speaker support, and natural delivery, designed for dialogue-driven applications, storytelling tools, and immersive voice experiences.

Online试用模型

Claude opus 4.7

LLM

Claude opus 4.7

$1.425 / M

Claude Opus 4.7 delivers a quantum leap in logical reasoning and creative synthesis. With its next-gen architecture and massive 2-million-token context, it serves as the ultimate high-performance partner for those pushing the boundaries of innovation and discovery.

Online试用模型

Gemini 3.1 Pro

LLM

Gemini 3.1 Pro

$0.5 / M

Gemini 3.1 Pro API is the latest general-purpose LLM developed by Google DeepMind, designed to bridge the gap between high-speed execution and deep logic. It empowers developers to build sophisticated agents with state-of-the-art accuracy in coding, creative writing, and cross-modal analysis.

Online试用模型

GPT 5.5

LLM

GPT 5.5

$0.14 / M

GPT-5.5 is OpenAI’s advanced reasoning model for agentic coding, knowledge work, scientific research, and complex multi-step task execution.

Online试用模型

Suno v5.5

AUD

Suno v5.5

$0.002 / req

Kie AI Music API is an AI music-generation model that converts text prompts into full vocal and instrumental tracks with natural dynamics and coherent musical progression, and it currently supports the latest V5.5 model for enhanced realism and scalable performance.

Online试用模型

Eleven labs

AUD

Eleven labs

$0.07 / 1K

ElevenLabs Eleven V3 enables expressive multilingual Text to Dialogue with audio tag control, multi-speaker support, and natural delivery, designed for dialogue-driven applications, storytelling tools, and immersive voice experiences.

Online试用模型

Claude opus 4.7

LLM

Claude opus 4.7

$1.425 / M

Claude Opus 4.7 delivers a quantum leap in logical reasoning and creative synthesis. With its next-gen architecture and massive 2-million-token context, it serves as the ultimate high-performance partner for those pushing the boundaries of innovation and discovery.

Online试用模型

Gemini 3.1 Pro

LLM

Gemini 3.1 Pro

$0.5 / M

Gemini 3.1 Pro API is the latest general-purpose LLM developed by Google DeepMind, designed to bridge the gap between high-speed execution and deep logic. It empowers developers to build sophisticated agents with state-of-the-art accuracy in coding, creative writing, and cross-modal analysis.

Online试用模型

GPT 5.5

LLM

GPT 5.5

$0.14 / M

GPT-5.5 is OpenAI’s advanced reasoning model for agentic coding, knowledge work, scientific research, and complex multi-step task execution.

Online试用模型

面向开发者的集成体验

通过统一、直观的 API，几分钟内即可完成接入。

STEP 01

选择模型

使用统一参数结构，从顶级媒体模型中选择，无需学习不同平台的 API。

VIDIMGAUDLLMVIDLLMIMGAUDIMGVIDLLMAUD

STEP 02

发送异步请求

提交生成任务。所有模型共用稳定可预测的 API 网关，切换模型非常轻松。

API

STEP 03

Webhook 与轮询

提供 webhook URL，在媒体生成完成时即时收到通知，或按架构需求使用标准状态轮询。

WEBHOOK

ReadyPOLL

为什么开发者选择 KIE

我们抽象掉多平台生成的复杂性，让你专注于构建自己的应用。

01

Lower-Cost Access

更低成本接入领先模型

KIE 通过灵活的积分系统提供更低成本的领先 AI 模型访问。相比官方 API 和其他市场，许多模型价格约低 30%，部分热门模型可节省 60–70%。

02

Unified API

一个 API 接入 100+ AI 模型

通过一个统一 API 访问视频、图像、音频和 LLM 模型。无需重建后端，即可在 Veo、Kling、Seedance、Runway、Claude、GPT、Gemini、Nano Banana、Suno 等模型间切换。

03

Reliability

99% 可用性，24/7 监控，稳定可靠

KIE 通过 API 可用性监控、全天候任务监控、异步任务追踪、webhook 回调和智能备用路由，保障高并发 AI 生成流程稳定运行。

04

Playground

Playground 中免费试用 AI API

在编写生产代码前，可直接在 Playground 测试 KIE API。尝试不同提示词、比较模型输出、调整参数，并在真实 API 环境中评估视频、图像、音频和 LLM 模型。

05

Drop-in

简单的 AI API 集成

一次集成即可访问视频、图像、音频和 LLM 领域的领先模型。清晰文档、统一任务流程和一致 API 模式，让团队只需修改 model_id 即可切换模型。

06

Security

可靠的数据安全

安全的 API 认证、加密请求和受控任务处理，帮助保护你的提示词、素材和生成结果。

07

Support

全年私有支持渠道

KIE 全年在线支持生产级 AI 工作流。API 用户可通过 Telegram 和 Discord 的专属私有渠道获得一对一支持，处理 API 错误、任务日志、webhook 回调、账单和集成问题。

技术团队看重 KIE 的什么

受到下一代 AI 原生应用构建者的信赖。

testimonials / 01

Production

我们已经基于 KIE API 部署了基础设施。对 Veo 和 Kling 生成使用 webhook 回调，而不是长轮询，为我们节省了大量服务器成本。

AI 视频 SaaS 创始人

常见问题

关于 KIE 集成与计费，你需要了解的一切。

01KIE.ai 是什么？basicsplatform

KIE.ai 是一个统一的 AI API 平台，让开发者可以通过一套一致的 API 接入领先的视频、图像、音频和 LLM 模型。你不需要逐一集成不同供应商，而是可以使用 KIE 以更简单的流程、更清晰的文档和更具竞争力的价格连接多个顶级 AI 模型。在许多情况下，KIE 的模型接入成本低于官方定价，通常低约 30%，部分模型甚至更具成本优势。 KIE 面向需要在生产环境中稳定、可扩展且高性价比接入 AI 模型的团队构建。

02KIE 上有哪些模型可用？models

KIE 支持多个类别的领先 AI 模型，包括视频、图像、音频和 LLM。视频生成方面，KIE 支持 Seedance 2.0、Veo、Kling、Grok Imagine 等模型。图像生成方面，KIE 支持 GPT Image 2、Nano Banana 以及其他高级图像模型。音频生成方面，KIE 支持 Suno、ElevenLabs 以及其他音频相关 API。 LLM 方面，KIE 支持 Claude、GPT、Gemini 等领先语言模型。你可以在模型市场查看完整模型列表：https://kie.ai/market

03如何查看模型价格？pricing

你可以在 KIE 定价页面查看当前模型价格：https://kie.ai/pricing 不同模型可能使用不同计费单位，例如按图片、按秒、按视频或按百万 tokens 计费。定价页面会展示每个模型适用的计费方式。

04KIE 提供折扣吗？pricingbilling

KIE 的价格公开透明。你可以在这里查看计费和充值详情：https://kie.ai/billing 对于较大金额充值，KIE 可能提供赠送 credits。例如，充值 $1,250 会额外获得 10% 赠送 credits，使实际使用成本约比标价低 10%。

05开发前可以在哪里测试模型？playgroundcredits

每个模型页面都包含 playground，你可以在集成 API 前测试模型。新用户还会获得 80 个免费 credits 用于测试，因此你可以在正式开发前尝试不同模型和工作流。

06在哪里可以找到 API 文档？docsintegration

你可以在这里查看 API 文档：https://docs.kie.ai/ 文档包含 API endpoints、请求示例、参数、callback 用法、响应格式以及特定模型的集成细节。

07任务失败会收费吗？billingtasks

不会。失败任务不收费。如果生成任务失败，KIE 不会扣除该失败任务的 credits。这能帮助开发者更放心地测试和运行生产工作负载，无需为未成功的生成付费。

08如果需要技术支持应该怎么做？support

KIE 为用户提供专门的技术支持渠道。你可以通过这个页面加入一对一支持群：https://kie.ai/vip-support 在那里你可以选择通过 Discord 或 Telegram 加入。我们的支持团队可以协助 API 集成、任务错误、callback 问题、计费问题和生产使用问题。

09KIE 是否足够稳定，可用于生产环境？reliability

是的。KIE 设计用于支持生产级用例和商业工作流。我们 24/7 监控模型可用性和任务性能，并持续优化路由和后端资源以提升稳定性。当出现问题时，团队会尽快定位并解决。

10KIE credits 会过期吗？creditsbilling

不会。KIE credits 不会过期。你的 credits 会保留在账户中，可随时使用。

11KIE 可以开具发票吗？可以添加公司信息吗？billing

可以。KIE 支持为企业用户开具发票。完成充值后，你可以在计费页面手动生成发票：https://kie.ai/billing 公司信息可以在开票流程中添加。

12如何查看已提交任务的结果？taskslogs

你可以在 Logs 中查看已提交任务和生成结果： https://kie.ai/logs Logs 页面可以查看任务记录、状态、参数、结果以及相关任务详情。

13在哪里可以查看最近的模型和功能更新？modelsupdates

登录后，你可以在 Updates 页面查看最近的模型发布、价格更新和功能变化： https://kie.ai/api-updates

14速率限制是多少？可以提高吗？rate limitsintegration

KIE 不对运行中的任务设置严格并发限制。对于新请求提交，默认限制是每 10 秒 20 个新请求，也就是每分钟最多 120 个新请求。这对大多数用户和生产用例已经足够。如果你收到 429 rate limit 错误，可以联系 KIE 支持申请更高的速率限制。我们会根据你的使用量和业务需求进行评估后调整。

15KIE 的文件和日志保留政策是什么？logsdata

媒体文件，包括图片、视频和音频，会保留 14 天。日志数据会保留 2 个月。超过这些保留周期后，数据将被永久删除。请提前下载并备份重要文件或任务日志。

立即使用全球顶级 AI 模型开始构建

在一个平台中访问视频、图像、音乐和聊天 API，更快、更实惠，也更适合开发者。选择 Veo 3.1 API、Runway Aleph API、Suno API 等模型，均具备稳定性能和简单集成体验。

获取免费 API Key