aid: vidu-ai url: https://raw.githubusercontent.com/api-evangelist/vidu-ai/refs/heads/main/apis.yml name: Vidu kind: company description: >- Vidu is a generative video AI platform from Shengshu Technology (ShengShu / 生数科技), built on the company's U-ViT diffusion-transformer architecture. The Vidu API turns text prompts, still images, and reference subjects into short video clips with features like text-to-video, image-to-video, reference-to-video (multi-entity / character consistency), start-and-end frame interpolation, and video upscaling. The API is a REST create-then-poll service - callers submit an asynchronous generation task, then poll the task status endpoint (or receive a callback) for the finished video URL. Access is open to individual developers and businesses on a prepaid, credit-based model with no application required. image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg tags: - Video Generation - AI Video - Generative AI - Text-to-Video - Image-to-Video - Reference-to-Video - U-ViT - Diffusion created: '2026-07-11' modified: '2026-07-11' specificationVersion: '0.19' apis: - aid: vidu-ai:vidu-ai-text-to-video-api name: Vidu Text-to-Video API tags: - Text-to-Video - Video Generation - Generative AI image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://platform.vidu.com/docs/text-to-video baseURL: https://api.vidu.com/ent/v2 properties: - url: https://platform.vidu.com/docs/text-to-video type: Documentation - url: https://docs.platform.vidu.com/7207545m0 type: APIReference - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Create a video from a text prompt alone. POST to /ent/v2/text2video with a model (viduq3-turbo, viduq3-pro, viduq2, viduq1), prompt, duration, resolution, and aspect ratio; the call returns a task_id in a created state. Poll the task creations endpoint (or supply a callback_url) to retrieve the finished video. Supports style, seed, movement amplitude, background music, audio-visual sync on q3, and off-peak processing. - aid: vidu-ai:vidu-ai-image-to-video-api name: Vidu Image-to-Video API tags: - Image-to-Video - Video Generation - Animation image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://platform.vidu.com/docs/image-to-video baseURL: https://api.vidu.com/ent/v2 properties: - url: https://platform.vidu.com/docs/image-to-video type: Documentation - url: https://docs.platform.vidu.com/7207545m0 type: APIReference - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Animate a single still image into a video clip. POST to /ent/v2/img2video with a model, one input image (URL or Base64; PNG/JPEG/JPG/WebP up to 50MB), an optional prompt, duration, and resolution. Returns an asynchronous task_id to poll for the generated video. Supports audio, background music, seed, movement amplitude, and off-peak mode. - aid: vidu-ai:vidu-ai-reference-to-video-api name: Vidu Reference-to-Video API tags: - Reference-to-Video - Character Consistency - Multi-Entity image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://platform.vidu.com/docs/reference-to-video baseURL: https://api.vidu.com/ent/v2 properties: - url: https://platform.vidu.com/docs/reference-to-video type: Documentation - url: https://docs.platform.vidu.com/334378351e0 type: APIReference - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Generate video that keeps supplied subjects visually consistent across the clip - Vidu's multi-entity consistency feature. POST to /ent/v2/reference2video with one to seven reference images plus a prompt, and the model composites those characters or objects into a new scene. Returns an asynchronous task_id to poll. Supports aspect ratio, resolution, background music, seed, movement amplitude, and off-peak mode. - aid: vidu-ai:vidu-ai-start-end-frame-api name: Vidu Start-End Frame API tags: - Start-End Frame - Interpolation - Video Generation image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://platform.vidu.com/docs/start-end-to-video baseURL: https://api.vidu.com/ent/v2 properties: - url: https://platform.vidu.com/docs/start-end-to-video type: Documentation - url: https://docs.platform.vidu.com/334381323e0 type: APIReference - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Generate a video that transitions between a supplied first frame and last frame. POST to /ent/v2/start-end2video with two images (start and end) and an optional prompt; the model interpolates the motion between them. Returns an asynchronous task_id to poll. Supports duration, resolution, seed, movement amplitude, and off-peak mode. - aid: vidu-ai:vidu-ai-upscale-api name: Vidu Upscale API tags: - Upscale - Super Resolution - Video Enhancement image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://docs.platform.vidu.com/7207545m0 baseURL: https://api.vidu.com/ent/v2 properties: - url: https://docs.platform.vidu.com/7207545m0 type: Documentation - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Upscale a previously generated Vidu video to a higher resolution (Upscale Pro). POST to /ent/v2/upscale referencing a prior creation or video asset; returns an asynchronous task_id to poll for the enhanced output. Request and response bodies here are modeled from the public Upscale Pro documentation and should be reconciled against the live reference on integration. - aid: vidu-ai:vidu-ai-task-query-api name: Vidu Task Query API tags: - Task Status - Polling - Async image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg humanURL: https://docs.platform.vidu.com/7207545m0 baseURL: https://api.vidu.com/ent/v2 properties: - url: https://docs.platform.vidu.com/7207545m0 type: Documentation - url: openapi/vidu-ai-openapi.yml type: OpenAPI - url: collections/vidu-ai.postman_collection.json type: PostmanCollection - url: collections/vidu-ai.opencollection.json type: OpenCollection description: >- Retrieve the status and results of an asynchronous generation task. GET /ent/v2/tasks/{id}/creations returns the task id, state (created, queueing, processing, success, failed), credits used, and a creations array of finished video URLs and cover images. This is the poll surface for every text2video, img2video, reference2video, start-end2video, and upscale task. Task listing and cancel-generation operations are part of the same task-management surface. common: - type: AgenticAccess url: agentic-access/vidu-ai-agentic-access.yml - type: Authentication url: authentication/vidu-ai-authentication.yml - type: GitHubOrganization url: https://github.com/shengshu-ai - type: LinkedIn url: https://www.linkedin.com/company/shengshu-technology - type: Website url: https://www.vidu.com - type: Documentation url: https://platform.vidu.com/docs - type: Plans url: plans/vidu-ai-plans-pricing.yml - type: RateLimits url: rate-limits/vidu-ai-rate-limits.yml - type: FinOps url: finops/vidu-ai-finops.yml maintainers: - FN: Kin Lane email: kin@apievangelist.com