Upload one image, describe the motion and get a 5, 10 or 15 second 720p or 1080p video, with optional synchronized audio.
Provider: Alibaba
Category: video generation
Endpoint: /v1/tasks
Status: Available
Cost: See live page pricing
Upload one image, describe the motion and get a 5, 10 or 15 second 720p or 1080p video, with optional synchronized audio.
Pricing
Standard Usage
Use images to control the first and last frame.
Ready for generation
Configure your parameters and click "Run" to see the output here.
Compare each input image with its clip; every example lists the full prompt and settings.
Default 720p, 5 seconds, no audio: the fox turns, its ears twitch and leaves drift down, matching the original light.
Prompt
The fox turns its head toward the camera, its ears twitch, golden leaves drift down through the misty light
A portrait image gives portrait video; with audio on, the wok, flames and crowd are generated with the picture.
Prompt
The vendor tosses the noodles in the wok, flames flare up, steam billows, the crowd chatters in the background
1080p output with a slow dolly around the car; sunset and dust detail stay crisp.
Prompt
The camera slowly dollies around the turquoise convertible as the sun sinks behind the mesas, dust drifting in the warm light
Ten seconds for the full move — tilting the umbrella, looking up, pushing in — in the image's 3:4 frame.
Prompt
She tilts the umbrella and looks up at the rain, raindrops run down the clear canopy, city lights shimmer in the puddles, the camera slowly pushes in
Including 1080p with audio and a 15-second take — tap to play.
Red fox at sunrise
Loom and glass city
Magnetic sand ripples
Fox trots across the clearing
Night-market service
Fox turns
Desert convertible
Choose by input and need:
Choose it when: You need a last frame or negative prompts
Trade-off: 5 or 10 s only
Choose it when: You want to set 4:3, 1:1 and other output frames
Trade-off: 720p only, up to 10 s
Choose it when: You need longer clips or first/last-frame control
Trade-off: More parameters to tune
Choose it when: You want a first/last-frame transition or an exact length from 2 to 15 seconds
Trade-off: Up to 15 s; 1-3 media items: first frame, last frame, driving audio or an opening clip
Choose it when: You need a 2-30 second take steered by first/last frames, or by reference images, video, audio, files or webpages
Trade-off: First/last frames and reference media can't be combined; pick one mode per task
Choose it when: You want a 2-30 second clip back faster, steered by first/last frames or by reference images, audio, files or webpages
Trade-off: No reference video input yet; first/last frames and reference media can't be combined
Current price: View live pricing
The per-second rate depends on resolution and the audio switch; see the pricing page for figures.
| Option | Good for | Relative price |
|---|---|---|
| 720p · silent (default) | Drafts, batches, your own soundtrack | Lowest |
| 720p · audio | Social clips that need atmosphere | About 2× the default |
| 1080p · silent | Finals and ads | About 1.5× the default |
| 1080p · audio | High-quality clips with sound | About 3× the default |
| Length | Fits | Example |
|---|---|---|
| 5 s | A single action or product reveal | Fox turns its head |
| 10 s | An action plus camera move | Umbrella tilt in the rain |
| 15 s | Scene mood, continuous action | Night-market service |
A clear, well-composed image works best; it sets the frame.
Describe action and camera, then choose length, resolution and audio.
Check the result in the Playground, then use the request below.
01
The image sets subject and scene; describe motion and camera.
Example:The fox turns its head toward the camera, leaves drift down
02
Push-in, orbit, tracking or handheld beats "cinematic".
Example:the camera slowly dollies around the car
03
For 10-15 seconds, list two or three actions in order.
Example:she tilts the umbrella, looks up, the camera pushes in
04
Name the sounds you want so the audio fits the picture.
Example:flames flare up, the crowd chatters in the background
curl --request POST 'https://api.hiapi.ai/v1/tasks' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"model": "wan2.6-flash/image-to-video",
"input": {
"prompt": "The fox turns its head toward the camera, its ears twitch, golden leaves drift down through the misty light",
"image_urls": [
"https://static.hiapi.ai/gallery/2026/10/ced2dd4214889094.jpg"
],
"duration": 5,
"resolution": "720p",
"audio": false
}
}'About Wan 2.6 Flash Image to Video
Wan 2.6 Flash Image to Video opens on your image and adds the action, camera move and mood from the prompt. The frame follows the image — landscape in, landscape out — with nothing extra to set.
Choose 5, 10 or 15 seconds at 720p or 1080p. Clips are silent by default; switch audio on for ambience and effects, billed at the audio rate.
Each block shows a clip and its settings: the frame follows the image, the prompt describes motion, camera and sound.
Wildlife
The input is a fox in a misty forest. The prompt asks it to turn toward the light, ears twitching, breath visible, as the camera pushes in.
With audio on, birdsong, leaves and a brook are generated in sync.
Mechanics and product
Complex machinery stays structurally stable: the shuttle races, threads ripple and a miniature city lifts as the camera orbits.
Wooden clacks and a metallic swish make the rhythm clear.
Abstract mood
Black sand ripples out from a copper ring as cool light sweeps across, filmed low and moving forward — good for openers and exhibition loops.
Audio is off here, billed at the silent rate and ready for your own music.
Long take
Fifteen seconds holds tossing, plating and handing over in one take — no stitching.
A vertical image gives vertical video for short-form feeds.
High-definition final
1080p keeps fur and leaf detail for large screens and ad finals.
Draft at 720p from the same image, then switch to 1080p for the final.
Subject, light and composition stay true to the input; only the described motion is added.
View this exampleSwitch audio on for sound that tracks the action — good for shots that need atmosphere.
View this example10 and 15 second clips hold a complete action and camera move without stitching.
View this example
Glances, turns and moving fabric for editorial clips.
Tossing, flames and steam with synced audio for atmosphere.
Natural, continuous animal motion and environmental detail.
Orbiting moves show the design; 1080p keeps the detail.
One API key for every video model; task history keeps clips and charges together for review and cost tracking.
One image and a prompt. The image opens the video; the prompt describes motion, camera and mood.
No. image_urls takes exactly one image.
JPG, PNG or WebP, at least 256×256 pixels and up to 10MB.
5, 10 or 15 seconds; 5 by default.
720p and 1080p; 720p by default.
No setting needed — the output follows the input image: landscape in, landscape out; portrait in, portrait out.
With audio on, the model adds ambience and effects synced to the picture. It is off by default, giving a silent clip.
Yes. Audio-on clips use the audio per-second rate, silent clips the silent rate, and 1080p costs more than 720p. See the pricing page for current prices.
Per second of video; the per-second rate depends on resolution and whether audio is on.
Yes. Chinese and English are both supported, up to 1,500 characters.
The image already sets subject and scene, so focus on action, camera and mood, e.g. "the fox turns toward the camera, golden leaves drift down, slow push-in".
Result links expire. Download or copy clips to your own storage if you need to keep them.