Animate a first frame into a 5- or 10-second video, with optional last-frame control.
Provider: Kuaishou
Category: video generation
Endpoint: /v1/tasks
Status: Available
Cost: See live page pricing
Animate a first frame into a 5- or 10-second video, with optional last-frame control.
Pricing
Standard Usage
Use images to control the first and last frame. With a last frame, use 16:9, 9:16 or 1:1.
Describe unwanted video content; Up to 2496 characters
Adjust adherence to the description; Range: 0 - 1
Ready for generation
Configure your parameters and click "Run" to see the output here.
Compare the first frame, last frame and video to see how character movement, position and colour changes build on the image’s composition.
A landscape first frame is cropped to 9:16 while the figure walks forward and neon reflects off the wet street.
Prompt
The woman in the red coat walks slowly forward under the clear umbrella, rain falling, neon signs flickering and reflecting on the wet street, steam drifting, smooth tracking shot
Input previews are center-cropped to the video shape. Open “Original” to see the full images.
The first frame places a red car on the left; the last frame places a blue car on the right. The clip moves between those positions and changes color.
Prompt
A toy car moves smoothly from the left to the right and changes from red to blue. Keep the simple light grey background and beige ground. Continuous static camera shot.
Choose by duration, resolution and input method:
Choose it when: You want adjustable 3–15 second clips or 1080p output
Trade-off: Takes a single first-frame image and has no aspect-ratio setting
Choose it when: You want an 8-second option
Trade-off: Fixed 720p, no last frame
Choose it when: You have a written idea without an image
Trade-off: Includes audio; no frame inputs
Current price: View live pricing
Set the opening image and add a last frame when you want to guide the ending.
Describe the action and camera, then choose 5 or 10 seconds and a ratio. Use 16:9, 9:16 or 1:1 for two frames.
Check the result in the Playground, then use the request below in your app.
curl --request POST 'https://api.hiapi.ai/v1/tasks' \
--header 'Authorization: Bearer YOUR_API_KEY' \
--header 'Content-Type: application/json' \
--data '{
"model": "kling-2.5-turbo/image-to-video",
"input": {
"prompt": "Waves crash against the rocks below, the lighthouse beam sweeps slowly across the storm clouds, mist drifts over the sea, slow cinematic push-in",
"image_urls": [
"https://static.hiapi.ai/gallery/2026/10/203945404507c091.jpg"
],
"duration": 5,
"aspect_ratio": "16:9"
}
}'KLING 2.5 TURBO / IMAGE-TO-VIDEO
Kling 2.5 Turbo Image-to-Video is the image-to-video version of Kuaishou's Kling 2.5 Turbo. Give it an image as the first frame, say what moves and how the camera travels, and get a 5- or 10-second clip. Add a last frame when the shot has to end on a specific picture.
First frame · last frame · length · ratio
The first frame is required, and it already settles what the subject looks like, where it stands and where the light falls. So the prompt skips the description and covers what happens next: where someone walks, how the mist moves, whether the camera pushes in or slides sideways. An optional last frame gives the motion somewhere to land.
Clips run 5 or 10 seconds. With a first frame alone you can pick 16:9, 9:16, 1:1, 4:3, 3:4 or 21:9; adding a last frame narrows that to 16:9, 9:16 and 1:1. Output is silent, which leaves music and voice-over to your edit.
Each clip below was generated for this page: a square product shot, a widescreen landscape, a landscape photo reframed as portrait, and a first-to-last-frame transition. Every one lists the first frame, the prompt, the ratio and length, and what actually changes on screen.
The first frame is a perfume bottle on a round marble stand with a few pink petals around it. Over ten seconds more petals drift in from the top and some settle on the stand. The camera swings a short way around the bottle, and the marble veining slides across to the other side.
The prompt also asked the bottle to rotate, and in the clip it barely does: the gold cap and cut glass stay exactly as photographed. If the product itself needs to turn, describe the turn more precisely, or add a last frame that shows it already turned.
1:1 · 1440×1440 · 10 s · first frame only
The perfume bottle rotates slowly on the marble stand, light glints across the glass, petals drift down, slow orbit
The first frame is a river valley at dawn, mist lying on the water and a wooden boat moored by the reeds. For five seconds the mist keeps moving. The camera shifts only slightly, yet the far ridge and the right-bank trees slide left while the boat edges right, and that offset gives the frame real depth.
The prompt asked for a slow pan, drifting mist and some foreground motion; it never mentioned the boat or the hills. 21:9 renders at 2200×940 and is only available without a last frame, which makes it a natural fit for title sequences and wide website headers.
21:9 · 2200×940 · 5 s · first frame only
Slow cinematic pan across the wide landscape, mist drifting, subtle motion in the foreground
The first frame is a 16:9 rainy alley, a woman in a red coat with a clear umbrella standing dead center with her back to us. With 9:16 selected, the clip keeps that middle strip: the figure, the umbrella and the neon signs on both sides stay, while most of the shopfronts at the edges are cut away.
The prompt asked for a slow walk forward, neon reflected on the wet street and a steady tracking shot. She walks deeper into the alley, puddles pick up pink and teal light, and steam rises on either side. When a landscape image is headed for a vertical feed, a centered subject is the safe choice.
9:16 · 1080×1920 · 5 s · first frame only
The woman in the red coat walks slowly forward under the clear umbrella, rain falling, neon signs flickering and reflecting on the wet street, steam drifting, smooth tracking shot
The first frame holds a single red toy car on the left; the last frame shows the same model in blue on the right. The red car rolls right, turns pale blue around the middle, deepens in color and stops exactly where the last frame placed it. The grey backdrop and beige floor never change.
The prompt spells out left to right, red to blue, an unchanged background and a locked-off camera. Because the two images differ only in the car, the model just fills in the movement between them. With both frames the ratio is limited to 16:9, 9:16 or 1:1; this one is 9:16.
9:16 · 1080×1920 · 5 s · first + last frame
A toy car moves smoothly from the left to the right and changes from red to blue. Keep the simple light grey background and beige ground. Continuous static camera shot.
Watch the clip: “From a red car to a blue car” in the examples above ↑
Turn a finished product shot into a 5- or 10-second clip. The first frame keeps the product as photographed; the prompt only handles petals, light and the camera.
Use a 21:9 landscape or interior photo as a widescreen establishing shot. Slow changes in mist, water or cloud are enough to carry five seconds.
Pick 9:16 for an existing landscape photo. Images with a centered subject come through the crop intact, ready for short-video platforms.
Pin the start and end with a first and last frame, so a move, color change or transformation lands on the picture you chose and cuts cleanly against its neighbours.
The whole picture comes from a prompt. Clips are also 5 or 10 seconds, in landscape, portrait or square, and every one comes with an audio track.
View model details →Pick any length from 3 to 15 seconds and choose 720p or 1080p. It takes a single first-frame image and has no aspect-ratio setting.
View model details →Clips run 5, 8 or 10 seconds in 16:9, 9:16, 4:3, 3:4 or 1:1, with the first frame center-cropped to fit. Output is fixed at 720p and there is no last frame.
View model details →Upload a first frame, say what moves and how the camera travels, and check five seconds first. Add a last frame when the ending has to be fixed, then send the same settings through the API.
Use jpeg, png or webp up to 10MB; jpeg and png are the safest choices. Try to match the image to the ratio you plan to output. When they differ, the clip may be center-cropped or may keep the image's own proportions, so keep the subject near the middle.
The last frame is an optional second image that decides where the clip ends. Keep the background and camera position the same in both images and change only what should change, like the red car turning blue on this page. With a last frame, choose 16:9, 9:16 or 1:1.
Pick 5 or 10 seconds. Ratios are 16:9, 9:16, 1:1, 4:3, 3:4 and 21:9, rendering at 1920×1080, 1080×1920, 1440×1440, 1660×1244, 1244×1660 and 2200×940. Resolution follows the ratio, so there is nothing else to set.
No. Clips come without an audio track, so add music or voice-over in your edit. If you want sound built in, Kling 2.5 Turbo Text-to-Video generates from text and every clip includes audio.
The first frame already shows the subject and setting, so describe motion, changes in the scene and the camera, for example “petals drift down, slow orbit”. You can also list things you don't want to see and adjust how closely the video follows the prompt.
Output links last about 7 days by default. Download anything you want to keep, or choose long-term storage when you create the task; past runs stay in your task history.
Use Image-to-Video when you have a picture and want the subject and composition to follow it. Use Text-to-Video when you only have an idea, or when the clip needs sound. Both run 5 or 10 seconds.
HiAPI Blog
