Wan v2.5
Wan 2.5 is an AI video generation model that creates cinematic 1080p videos up to 10 seconds long with synchronized audio, realistic motion, and consistent character and environmental details. It excels at understanding complex creative prompts to deliver professional-grade camera movements, natural physics simulation, and seamless lip-sync capabilities across multiple languages.
Available for Text and Image to Video.
Choose your Generation Mode
Text to Video
Script to Screen.
Describe a scene, camera movement, or action in text, and the AI will generate a video clip from scratch.
Image to Video
Bring Images to Life.
Upload a static image and have the AI animate it. Perfect for adding movement to photos or art.
Why use Wan v2.5?
Audio-Synced Generation
Revolutionary lip-sync technology that matches character mouth movements to audio input with high precision, including voices, sounds, and music.
T2V & I2V Modes
Generate videos from text prompts (T2V) or animate still images (I2V) with full control and consistency.
High-Resolution Output
Supports 480p, 720p, and 1080p at 24 FPS for cinematic-quality videos up to 10 seconds.
Try These Prompts with Wan v2.5
"A dimly lit jazz bar at night, wooden tables glowing under warm pendant lights, patrons sipping drinks and chatting quietly, a three-piece band performs on stage with the saxophone player under a spotlight, camera slow pan across the crowd then gentle zoom to the sax player's expressive hands during solo, ambient smooth live jazz music with saxophone, piano, clinking glasses, low murmurs and occasional laughter, no dialogue."
Creates an atmospheric jazz performance scene with synchronized ambient audio and smooth camera motion.
"The white dragon warrior stands still in a grand cathedral-like structure with towering stone arches, glowing golden eyes fixed forward in determination, camera slowly circles around revealing intricate white scale armor with gold accents, maintains strong heroic posture, soft choral tones and echoing ambient sounds build majestic tension, ending in close-up on fierce face."
Produces a heroic character reveal with detailed armor, circling camera, and immersive choral audio.
"A young detective in a rainy noir city street at midnight, trench coat flapping in wind, approaches a shadowy informant by a flickering neon sign, detective says "You promised answers, now talk before it's too late" in gravelly urgent tone, informant whispers nervously "They're closer than you think", camera dolly zoom from wide street view tightening on their tense faces with lip-sync, thunder rumbles and rain patters, moody blue lighting."
Generates synchronized lip-synced dialogue exchange in a classic noir style with dynamic weather audio and camera zoom.
"Golden hour over misty mountain valley, ancient pine trees sway gently in breeze, a crystal-clear river flows downstream carving through rocks, wild deer grazes peacefully at water's edge then looks up curiously, camera slow aerial drift from wide panoramic vista down to ground level following river flow, ambient birdsong, wind through leaves, and gentle water rushing sounds."
Achieves tranquil landscape motion with natural animal animation, drifting camera, and layered environmental audio.
Sample prompts — click any card to copy
How to generate
Go to Tool
Select "Text to Video" or "Image to Video" above.
Select Model
Ensure Wan v2.5 is selected in the dropdown.
Enter Script
Describe the motion, camera angle, and subject clearly.
Generate
Processing takes longer than images. Be patient!
Compare Video Models
Not sure if Wan v2.5 is the best for your clip? Compare it against others in the Video Playground.
Open Video PlaygroundMade with ❤ by AI4Chat