MiniMax H3 First to Last Frame: Animate Between Two Images
Last verified: November 1, 2026
MiniMax H3 is best used for shots where the ending is already decided. Think of a product that has to stop on the approved hero angle, a room that has to finish in its renovated state, or a landscape that has to close on one exact composition. With MiniMax H3 First to Last Frame you supply the opening still and the closing still, and the model generates one continuous movement from the first to the second. The prompt guides that movement: it describes the path between the two pictures, not the pictures themselves.
Each still has a set role in the generation: one is the opening frame and the other is the ending frame. First-and-last-frame input is one of the modes MiniMax H3 supports at both of its resolution tiers and across its whole duration range. That is the difference from animating a single image. You choose the ending as a picture, so the model doesn't have to invent one.
Vidofy's form works the same way. Two required upload slots take your opening and closing stills, the prompt box takes your description of the movement, and two settings set the clip's length and how sharp it renders. Your images fix both ends, so the prompt does most of the creative work. That is why the tips and fixes further down focus on writing the middle of the shot.
Explore more first to last frame
Inputs and Output at a Glance
What a run takes in (two stills, a prompt and two settings) and what you get back.
Input
First Frame (JPG, JPEG, PNG or WEBP, up to 10 MB) + Last Frame (JPG, JPEG, PNG or WEBP, up to 10 MB) + a text prompt of up to 7,000 characters
Video Duration
4 to 15 seconds
Resolution
768P or 2K
Output
1 video per run
Tips for Better First-to-Last-Frame Results
Write the Movement as Four Beats
Match the Described Timeline to the Video Duration
Name Concrete Changes and Camera Moves
How to Make a Video From a First and Last Frame
Six steps take you from two stills to a downloaded clip.
Upload the Opening Still in First Frame
Click "Upload first frame" and add the image the clip should start on. The "First Frame" slot is required. Pick a sharp, well-lit still with the subject placed where the motion should begin: this exact picture opens the video, so any clutter or softness in it carries into the clip.
Upload the Closing Still in Last Frame
Click "Upload last frame" and add the image the clip should end on. "Last Frame" is required too. Choose an ending that one continuous movement could plausibly reach from your opening still, such as the same subject or space in a new state, angle or light. That gives the prompt a believable route to describe.
Describe the Path in the Prompt Box
The "Prompt" field is required and holds up to 7,000 characters, far more than one transition needs. Spend the words on the change between your two stills, not on describing them again. If you only have a rough idea, the "Enhance Prompt" button under the box suggests fuller versions. Your text stays as you wrote it unless you pick one of the suggestions.
Choose Video Duration and Resolution
"Video Duration" goes from 4 to 15 seconds in one-second steps and defaults to 4. "Resolution" offers 768P (the default) or 2K. Make drafts at 768P with a short duration to check that the movement works. Then re-run the same frames and prompt at 2K for the version you deliver, and keep the duration you settled on so the pacing stays the same.
Start the Render With Generate
Click "Generate" once both stills are in their slots and the prompt is ready. Before you click, check the prompt's last line against your closing image. If they describe different things, fix the text first, because the model is working toward that picture.
Download the Finished Clip
Each run returns one video. Download it and compare its final second with your last-frame image. If the path is right but the timing feels off, change "Video Duration" and run again. If the route itself is wrong, rewrite the middle of the prompt instead.
First-to-Last-Frame or Another MiniMax H3 Workflow?
Three other ways to work with MiniMax H3, and when each fits better than two fixed frames.
| Criterion | MiniMax H3 | Alternatives | Best For |
|---|---|---|---|
| Control over how the shot ends | You provide both the opening and the closing still, so the shot is planned from the start to finish on the image you put in Last Frame. | The Image To Video workflow anchors only the opening still and develops the shot forward from it, so the ending stays open. | Pick first-to-last-frame when a brief, storyboard or approved packshot fixes the ending, and pick Image To Video when you want to see where the model takes a single still. |
| What you need before you start | Each run needs two prepared stills plus a prompt that describes the path between them. | The Text To Video workflow builds the whole timeline from the written prompt alone. | Use Text To Video for early ideas before any frames exist, then move to first-to-last-frame once you have stills for the opening and the ending. |
| How your images are used | Your two images become the clip's actual first and last frames, with the prompt filling the movement between them. | The Reference To Video workflow takes up to nine images, three video clips and three audio clips as references. It is aimed at consistent characters, motion reference and clip continuation. | Stay with first-to-last-frame when the exact opening and closing pictures matter, and choose Reference To Video when a character or product has to stay recognizable in framing you don't want to fix in advance. |
Common First-to-Last-Frame Problems and Fixes
What usually goes wrong when connecting two stills, and what to change before the next run.
The clip cuts to a different shot halfway instead of flowing into my last frame
The movement plays backwards from what I planned
My frame image is rejected or the run won't go through
The prompt box won't take any more text
Frequently Asked Questions
Which image files can I upload as the first and last frame?
The First Frame and Last Frame slots accept the same files: JPG, JPEG, PNG or WEBP, up to 10 MB per image and at least 300×300 pixels. If you plan to render at the higher resolution setting, export your stills well above that minimum.
Can I make a video from just one image instead of two?
Not in this workflow. Every clip here is built from an opening still and a closing still, and both slots have to be filled. If only the opening matters and the ending can stay open, use the Image To Video workflow for MiniMax H3, which starts from a single still.
What makes one generation cost more than another?
The credit cost of a run is calculated from the options you choose. Once you're signed in, it appears above the Generate button. Changing Video Duration or Resolution changes the figure, so check it there before each run. The Enhance Prompt button doesn't use credits.
How do I build a sequence longer than one clip?
Plan it as a chain of runs. The still you used as the Last Frame of one run is the image that clip is built to end on. Upload that same file as the First Frame of the next run and add a new closing still, so each clip starts where the previous one stopped. Use the same style wording in every prompt to keep lighting and color grade consistent across the sequence.