MiniMax H3 Lite | Image to Video.

MiniMax H3 Lite | Image to Video Overview

MiniMax H3 Lite | Image to Video turns a starting image and a text prompt into a short generated video. Created by MiniMax within the H3 family, this workflow anchors the opening with a required first-frame image and optionally accepts a final-frame image. The Lite integration offers 480P and 768P, so you can choose a lower-resolution draft or a larger output.

The prompt describes the desired action, scene development or camera direction while the supplied image establishes the visual starting point. Output durations range from 4 to 15 seconds, so the workflow is intended for individual shots and short transitions.

On each::labs, this MiniMax video generation workflow helps creators explore motion from existing visual assets. Use it to develop footage for review and editing, while checking that the generated movement and ending support the intended composition.

Capabilities

• Generate a short moving clip from a required first-frame image and a written motion instruction.
• Accept an optional final-frame image when the creative brief includes a target ending composition.
• Request whole-second output durations across the supported range from 4 to 15 seconds.
• Choose 480P or 768P output from the settings currently available in this integration.
• Derive the output aspect ratio from the image rather than requiring a separate ratio selection.
• Interpret prompt descriptions of subject movement, environmental action or camera direction as generation guidance.
• Return completed footage as a video URL for visual review and subsequent editing workflows.

Use Cases for MiniMax H3 Lite | Image to Video

MiniMax H3 Lite | Image to Video can help creators develop short motion studies from images they already have. A social creator might animate a still landscape with “Gentle waves roll toward the shore while the camera remains steady,” then select a take for a larger edit.

A marketer can explore motion around an existing product photograph, using a prompt such as “A slow camera move reveals the object under soft light.” Review the object’s appearance before treating the result as finished advertising.

A designer can supply starting and ending compositions to explore a brief visual transition between two approved concepts.

A developer can expose the first-frame and optional last-frame fields in an image-animation tool. These scenarios share the same starting point: an image establishes the shot before motion is generated.

Tips and Tricks

Describe the change you want instead of rewriting every visible detail in the source. Try: “The fabric moves gently in the breeze while the camera slowly moves closer; keep the framing centered on the garment.” This gives the animation a clear focus without several competing actions.

For a landscape image, a second starting point is: “Clouds drift across the distant hills as sunlight brightens the valley; keep the camera steady.” These prompts are suggestions for iteration rather than tested guarantees.

If adding a last frame, choose images with compatible framing and make the requested transition explicit. Compare one change at a time, retaining the same source while revising the motion instruction. Start with a short duration, review the generated movement, then extend the shot only if the action needs additional time.

Technical Specifications

The MiniMax H3 Lite | Image to Video API accepts these inputs and returns generated video:

• Prompt: required, nonempty text of up to 7000 characters describing the desired result.
• First frame: required image_url pointing to the starting image.
• Last frame: optional end_image_url to provide a target ending image.
• Duration: integer seconds from 4 to 15; the default is 5.
• Resolution: 480P or 768P, with 768P selected by default.
• Aspect ratio: determined by the source image; no separate ratio parameter is exposed.
• Output: video URL after successful asynchronous generation; no output codec or frame-rate setting is exposed.

Things to Be Aware Of

A reachable URL is not enough if the underlying image violates the provider’s media requirements. Images must be supported formats, no larger than 30 MB, with width and height between 256 and 5760 pixels. The permitted width-to-height ratio is 0.4 to 2.5.

A very different ending composition may make the requested transition difficult to communicate within a short clip. Inspect changes in fine detail and subject appearance rather than assuming the starting image stays unchanged throughout.

Wait for successful completion before using the output video.

Key Considerations

Choose a source image with a clear subject and enough room for the movement you want to explore. If supplying a last frame, plan a transition that can reasonably fit the requested duration. The ending image guides the request; review the actual result.

The catalog price is $0.025 per requested output second at 480P or $0.04 at 768P. A five-second clip costs $0.125 or $0.20, respectively.

Use images you are permitted to submit, and make their URLs reachable for processing. Check the completed clip for visual continuity before including it in a published edit or campaign.

Limitations

This integration requires a first-frame image; an ending image alone is not a complete request. It does not expose reference-video, reference-audio or reference-image arrays in this mode. Output duration is limited to 4–15 seconds, and 2K is unavailable.

No seed, negative-prompt, output frame-rate or codec control is exposed. A supplied image does not guarantee that every detail, face, label or object shape remains exact throughout generated motion.