Generate high-quality videos with Aliyun Wan 2.7 and Wan 2.6 models. Async DashScope passthrough with direct Aliyun CDN output URLs.
Auto-expand the prompt for better quality (recommended).
Add an Aliyun AI-generated watermark to the output.
Gacha mode: submit multiple tasks with the same settings
Default per-second price. Failed tasks are not charged.
No videos generated yet
Wan 2.7 and Wan 2.6 are Aliyun's video generation models, accessed via DashScope passthrough. One endpoint, one schema — switch capability by changing the model name.
Turn a prompt into a 5–10s cinematic clip. prompt_extend auto-enriches your description for better quality.
Animate a first-frame image. Wan 2.7 i2v additionally supports driving audio for lip-sync and rap.
Supply up to 9 reference images and the model preserves subject, clothing, and scene characteristics.
Edit a source video with natural-language instructions and up to 5 reference images — swap clothing, backgrounds, and more.
Aliyun-grade video quality through a clean async passthrough, with output that streams straight from Aliyun's CDN.
Wan 2.7 and Wan 2.6 variants, covering every video capability through one DashScope passthrough endpoint.
Text-to-video. Prompt-only generation at 720P / 1080P, 5s or 10s.
Image-to-video from a first-frame image. Wan 2.7 i2v also accepts driving audio.
Reference-to-video with up to 9 reference images, preserving subject and style.
Video editing from a source video plus up to 5 reference images via natural-language instructions.
720P and 1080P output. Higher resolutions and longer clips take more time.
Uses a dedicated APIYI Wan-group token, configured separately in Settings.
Have questions? Contact us at hi@apiyi.com for more help.
Generate Aliyun-quality videos across four capabilities with direct CDN output.