Use a first frame, last frame or both to guide a clip’s opening and ending, then describe motion and sound.
Provider: MiniMax
Category: video generation
Endpoint: /v1/tasks
Status: Available
Cost: See live page pricing
Use a first frame, last frame or both to guide a clip’s opening and ending, then describe motion and sound.
Pricing
Standard Usage
Set the opening frame and canvas ratio
Set the ending frame. With only a last frame, its image sets the canvas ratio
Range: 0.92 - 15
Supply a first or last frame. The first frame sets the canvas; with only a last frame, that image sets it. No aspect ratio is submitted.
Replace the output soundtrack with audio of at least 2 seconds and at most 15 MB. Longer audio is trimmed; shorter audio is padded with silence. Video duration stays unchanged
disabled skips expansion; balanced is the default; quality adds detail and may take longer
Ready for generation
Configure your parameters and click "Run" to see the output here.
Compare the input images and clips to see the evolving water and how endpoint frames arrange the boat’s position.
The first frame supplies a flat illustration of the bridge, red boat and banks. The clip builds on that composition with movement in the water and raindrops.
Prompt
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
A last frame alone guides the shot toward the ending composition with the boat on the right. The portrait illustration keeps the arch and banks, with the supplied target audio replacing the soundtrack.
Prompt
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
Target audio input
The first frame places the boat on the left and the last frame sets its position on the right. The clip connects these compositions within the same bridge illustration, keeping its simple colors and portrait layout.
Prompt
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
A square product photograph anchors the first frame while an orbiting camera reveals the woven upper, laces and sole, with soft light and a shifting contact shadow. The clip includes a generated soundtrack that can be auditioned before choosing music for the product presentation.
Prompt
The camera slowly orbits the sneaker on a white studio background, soft light sweeping across the fabric, subtle shadow shifting beneath it. No text or subtitles.
The portrait illustration keeps the layered mountains, lake and cabin while drifting clouds and rippling water add gentle motion. Warm window light accompanies the deepening dusk, and a generated soundtrack makes the clip a starting point for a calm opening background.
Prompt
Clouds drift slowly across the sky, the lake surface ripples gently, warm cabin windows begin to glow as dusk deepens. No text or subtitles.
Stay here when you have pictures to work from. If you only have a written idea, the Text-to-Video entry fits better.
Current price: View live pricing
Supply at least one endpoint image. The canvas follows the first frame when present, otherwise the last frame.
Direct the action, camera and sound, then choose length and resolution. Do not set aspect_ratio.
Review the opening and ending, then use the request below with the same inputs 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": "minimax-h3-max/image-to-video",
"input": {
"prompt": "A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.",
"resolution": "480P",
"duration": 1.5,
"first_frame_image": "https://static.hiapi.ai/model-onboarding/minimax-h3-max/2026/10/13/1791862423659-cd76600c6f3d1b5d-first-frame.png",
"seed": 0,
"prompt_expansion_mode": "disabled"
}
}'MINIMAX / H3 MAX · IMAGE-TO-VIDEO
Upload a first frame to set the opening, a last frame to set the landing, or both, and describe only the motion in between. The picture fixes subject and composition, so products, characters and brand looks stay intact.
MODEL OVERVIEW
Image-to-video starts from a picture you already have: a product shot on white, an illustration, a poster or one storyboard panel. Since the image sets subject and composition, results are more controllable than text alone and the look holds.
First and last frames work alone or together; with a single image, the video takes that image's shape. JPG, PNG and WebP up to 30 MB each are accepted, and target audio can replace the soundtrack.
Each example shows the input frame above the result; expand to read the prompt.
The first frame is a flat illustration of an arch, a red boat and the bank. The clip keeps its color blocks and vertical layout and animates only the water and rain.
With an existing poster, illustration or product shot, keep the composition clear and describe only what should move, so the original design stays intact.
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
Only a last frame was supplied, so the camera moves toward the ending composition with the red boat on the right, and the image sets the arch and bank.
Use it when the closing image is decided and you want a clean hand-off to the next edit. This clip's soundtrack comes from uploaded target audio.
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
The first frame puts the boat on the left and the last frame on the right; the video connects the two within the same illustration.
Keep subject, scene and shape consistent across the two frames and allow time for the move. It suits simple transitions and products changing position.
A red paper boat drifts beneath a stone bridge after rain. Smooth tracking camera, natural ripples, soft birdsong. No text or subtitles.
A square product shot is the first frame. The orbiting camera reveals the knit upper, laces and sole outline while soft light and contact shadows shift with the angle.
It fits product pages and e-commerce hero videos. The clip includes a generated soundtrack that you can keep or replace with your own music.
The camera slowly orbits the sneaker on a white studio background, soft light sweeping across the fabric, subtle shadow shifting beneath it. No text or subtitles.
The vertical illustration keeps its layers of hills, lake and cabin while clouds and ripples move gently and the warm window light deepens with dusk.
It turns a static poster into a calm opening background or event-page loop, with the original composition and palette preserved.
Clouds drift slowly across the sky, the lake surface ripples gently, warm cabin windows begin to glow as dusk deepens. No text or subtitles.
Use a product shot on white as the first frame and orbit or push in to show material and detail on product pages.
Add motion only to water, weather or light and keep the rest still for covers, event pages and social posts.
Lock the closing composition with a last frame so the clip hands off cleanly to the next one.
Keep both frames and change only the camera and pacing in the prompt to compare pans, push-ins and orbits.
Upload a first or last frame in the Playground and try a clip with one line of motion. The images, prompt and result of every run stay in task history, so the next round can change a single variable.
Supply at least one image. A first frame sets the opening, a last frame alone guides the ending, and both define the opening and ending compositions. These are video endpoint frames, not general style references.
The first frame sets the canvas when present; otherwise the last frame does. This entry does not accept aspect_ratio. Prepare a landscape or portrait image for your destination and preferably match both frame shapes. The video determines the final pixel dimensions.
The image already supplies the subject and composition. Focus the prompt on action, environmental changes and how the camera continues. To guide an ending position, use a last frame with a motion direction and allow enough time for the action.
Yes. A last frame used alone is both the ending picture and the frame shape for the whole clip. Describe the action that leads from the opening to that ending in the prompt.
The first frame sets the frame shape and the last frame is handled to match, but a big difference can make the transition feel awkward. Use the same orientation and similar compositions for both.
No. On this entry the images are the video's start and end frames, not general style references. If you want to start from text, use the text-to-video entry.
Send target_audio_url and the finished soundtrack is replaced with your audio. The audio must be at least 2 seconds and no more than 15 MB, and the video at least 2 seconds. It does not guarantee that the picture follows the sound, so describe the action in the prompt.
1080P is an output tier refined from a 768P source. In our test with two portrait frames it returned 1080×1920. Use it for final delivery and use 480P when comparing prompts.
Duration accepts fractions from 0.92 to 15 seconds. A 1.5–2 second test confirms that the frame was used correctly and the orientation is what you expect before you scale up to the final length.
HiAPI Blog
