Text to video
Choose the supported inputs and model-aware controls in one focused AllVideoAI workflow.
Alibaba · Video model
Create with Wan 2.7Wan 2.7
Flexible text, image, and source-video generation with optional audio guidance.
Model overview
Meet Wan 2.7
Wan 2.7 is an Alibaba model family exposed through WaveSpeedAI for text, image, reference-video, editing, and extension workflows. Its text mode is positioned around coherent cinematic clips, stable motion, crisp detail, and strong instruction following.
Reference-to-video can combine video references and optional images to preserve character, style, or visual identity in a new scene. It is a practical choice when a project needs several creation paths under one family.
Image to video
Choose the supported inputs and model-aware controls in one focused AllVideoAI workflow.
Video to video
Choose the supported inputs and model-aware controls in one focused AllVideoAI workflow.
Prompting tip
For references, call out each source explicitly and state which identity, motion, style, or composition should be borrowed from it.
AllVideoAI Studio
Compare models and create in one workspace
Supported creation modes and parameters adapt to the selected model, with credit pricing confirmed before generation.