Video generation
Text-to-video and image-to-video tasks, including reference video, head-tail frames, multi-frame video, video extension, lip-sync, image generation, and video-to-video.
Core routes: /v2/aigc/text_to_video, /v2/aigc/image_to_video
Back to products
VOD AIGC API Layer
VIDEO on Demand AIGC
A focused set of atomic AIGC capabilities for video generation, multimodal reasoning, quality enhancement, smart highlights, and video translation.
6
AIGC atoms
65%
HappyHorse1.1 discount
POST
VOD API tasks
Overview
Only the AIGC capability surface.
This page narrows VIDEO on Demand down to the AIGC and AI-processing APIs customers need for creative workflows.
Model discounts
Commercial discount matrix.
Discounted price uses the official list price multiplied by the coefficient below.
| Model | Discount | Discounted Price Description |
|---|---|---|
| Vidu | 60%40% off | Official list price x 0.6 |
| PixVerse V6 and earlier versions | 55%45% off | Official list price x 0.55 |
| PixVerse V6 and C1 | 75%25% off | Official list price x 0.75 |
| Keling | 65%35% off | Official list price x 0.65 |
| Liblib | 65%35% off | Official list price x 0.65 |
| HappyHorse1.1 | 65%35% off | Official list price x 0.65 |
| Hailuo | 60%40% off | Official list price x 0.6 |
| BG (nano banana, veo, G models) | 70%30% off | Official list price x 0.7 |
| GPT-Image2 | 70%30% off | Official list price x 0.7 |
AIGC capabilities
Five atoms, no extra VOD noise.
Multimodal model chat
OpenAI-compatible chat completions for G3FP, G3PP, and G31PP, with non-streaming responses and optional thinking configuration.
Core route: /v2/chat/completionsZhigan Ultra HD
AI quality rebirth for enhancement, super resolution, HDR, frame interpolation, restoration, face enhancement, denoise, and audio normalization.
Core route: /v2/medias/process via workflow or presetSmart highlights
Generate one-click films, highlight clips, narration videos, precision edits, and single-episode summaries after video analysis.
Core route: /v2/highlight/job/narrationVideo translation
Create project-based subtitle and speech translation workflows with OCR or ASR recognition, subtitle erasure, AI dubbing, voice clone, MP4 output, and editable project export.
Core routes: /v2/translation/project, /v2/translation/tasks
Video translation flow
Project -> Task -> Output
Use a translation project to group multiple tasks before creating media-level translation jobs.
POST /v2/translation/project
Submit one or more media IDs with source language, target language, subtitle mode, and speech mode.
POST /v2/translation/tasks
Read task status and output URLs, including final MP4, cover, erased-subtitle video, and intermediate artifacts.
GET /v2/translation/project/{projectId}/tasks
Export completed tasks as an editable project package for downstream timeline and subtitle editing.
POST /v2/translation/task/{taskId}/export_project
Subtitle
OCR / ASR
Recognize burnt-in subtitles or speech text, then produce translated subtitles with optional hard-subtitle composition.
Speech
AI dub / voice clone
Generate translated speech through preset AI voices or speaker-tone cloning for character-preserving output.
Erase
Global / dialog / manual
Use v3 or v4 subtitle erasure and choose full-video, dialogue-area, or custom pixel-region removal.
Languages
15+ target locales
Subtitle translation covers major global languages, with AI dubbing and voice clone available on matched language ranges.
API methods
Choose an API and copy the call shape.
VOD uses BCE authorization headers. Generated tasks return task IDs that can be queried later through task-detail APIs.
{
"model": "K26",
"modelK26TaskInput": {
"duration": 5,
"prompt": "A cinematic product video with dynamic camera movement",
"negativePrompt": "blur, low quality",
"cfgScale": 0.5,
"aspectRatio": "16:9",
"mode": "pro",
"sound": "on"
}
}
Pricing
Post-paid AIGC pricing reference.
Billing is post-paid. Items are metered separately and no usage means no charge.
Text-to-video, image-to-video, reference video, and head-tail-frame video vary by duration and resolution. Multimodal model listed range: CNY 0.36-21.6/sec.
Quality enhancement starts at CNY 0.15/min. Super resolution, HDR, frame interpolation, restoration, face enhancement, and audio tools are priced by output profile.
Video analysis: CNY 1.6/min. One-click film and narration: CNY 0.8/min. Highlight clip and precision edit: CNY 0.5/min.
Subtitle translation: CNY 2/min. AI dubbing: CNY 1/min. Voice clone: CNY 4/min. Subtitle erasure can add CNY 1-2.5/min.
| Capability | Billing unit | Official list price reference |
|---|---|---|
| Text-to-video | Task | CNY 0.16-1.41 by duration and resolution |
| Image-to-video | Task or second | CNY 0.31-5.94 by duration and resolution |
| Smart highlights | Input video minute | CNY 0.5-0.8/min after analysis |
| Video translation | Output video minute | CNY 1-4/min by subtitle, AI dubbing, or voice clone |
| Zhigan Ultra HD | Processed minute | CNY 0.15-6/min by enhancement operator and resolution |
Test console
Build and test signed VOD API requests.
Paste a BCE Authorization header generated by your backend or the official API debug console. This page does not store credentials.
References