Multimodal video generation
The API is presented as a platform for advanced multimodal models, with documented video workflows based on text, images, keyframes, and reference inputs.
Vidu API is a developer platform for integrating AI video generation into applications. It provides API workflows for text-, image-, keyframe-, and reference-based video creation, along with access to other Vidu model capabilities.
Vidu API is a developer-focused platform for integrating Vidu’s multimodal video-generation models into applications. It supports documented workflows for creating video from text, images, keyframes, and references, with additional platform options for image-to-video, start/end-to-video, and real-time video.
The service includes API documentation, a quickstart, an API Playground, model and pricing information, and credit-based account billing. It is intended for developers and enterprises building or scaling video-driven products rather than only using a standalone video editor.
The API is presented as a platform for advanced multimodal models, with documented video workflows based on text, images, keyframes, and reference inputs.
The text-to-video documentation describes JSON request data and token-based authorization for submitting video-generation requests.
Platform navigation lists image-to-video, start/end-to-video, reference-to-video, and real-time video alongside text-to-video.
The platform organizes capabilities across video generation models including Vidu Q3, Q2, and Q1, with separate categories for image, audio, avatar, and editing models.
Accounts use rechargeable credits, while plans specify validity periods and concurrent-job allowances. Tasks above a plan’s concurrency limit enter a queue.
A quickstart, API documentation, API Playground, pricing console, and Help Center provide resources for implementation and evaluation.
Developers can use the API documentation and quickstart to connect Vidu generation workflows to a video-driven product or application.
Applications that start with prompts, images, keyframes, or reference material can select the corresponding generation workflow instead of using a single text-only process.
Teams can review the model catalog, use the API Playground, and compare credit consumption by model, resolution, task type, and duration.
Creators, studios, and professional content teams can use higher-volume credit packages with stated concurrency allowances for ongoing generation workloads.
Products that require more immediate or interactive output can investigate the platform’s listed real-time video capability, subject to the current API documentation and account limits.
The platform is described for enterprises and developers building and scaling video-driven products with Vidu’s multimodal models.
The platform lists text, images, keyframes, and reference inputs, as well as image-to-video, start/end-to-video, and real-time video workflows. Specific input requirements depend on the selected API operation.
The text-to-video documentation shows an Authorization header using a token and JSON as the data exchange format. Consult the current API documentation for the exact request and credential setup.
Usage is based on rechargeable credits. The pricing page states a standard rate of $0.005 per credit, while actual consumption varies by capability, model, resolution, task type, and duration. Credits have plan-specific validity periods.
The pricing page states that tasks exceeding a plan’s fixed concurrent-job allowance automatically enter a queue. The allowance varies by recharge plan.
segmind.com
Segmind 通过 API、Pixelflow 模板和模型访问,自动化图像与视频工作流,并提供个人、团队及企业套餐。
thinkdiffusion.com
用于开源 AI 图像和视频工具的云端工作区
producer.ai
Google Flow Music 是一个生成式 AI 平台,可用于创作、混音和分享歌曲,并在支持免费与付费方案的网页应用中提供音乐视频生成、音频混音和工具构建功能。
onorca.dev
Orca 是用于借助编码代理交付产品的智能体开发环境,支持在隔离工作树中并行运行多个 CLI 代理,并提供桌面端和移动端协同工作流。
rizzle.com
Rizzle 将文章转化为编辑级视频,提供 AI 辅助与人工审核,帮助出版商和企业进行多渠道分发与变现。
easysite.ai
EZsite AI 是一款 AI 网站构建工具,可将 URL 转换为完整的 React 或 Vue.js 应用,并提供托管、自定义域名、代码导出及面向后端的功能,帮助团队交付可部署项目。