Seedance 2.0 Mini First to Last Frame: Two Images to Video

Last verified: November 3, 2026

Seedance 2.0 Mini First to Last Frame suits production shots where you already know both ends. Sealed packaging finishes unboxed. An empty room finishes furnished. A valley at dawn finishes in full sun. You can frame these as ultra-wide 21:9 footage, or as standard or vertical clips. ByteDance built Seedance 2.0 Mini for everyday production workflows. Its first-and-last-frame generation works from a pair of images: one opens the clip and one closes it.

The form follows the same idea. The opening image goes into First Frame and the closing image into Last Frame. The prompt describes what happens in between: the action, the camera move and the pace. All three are required, so every run starts with a set start state, a set end state and a written plan for the move that joins them. Next to them you'll find length, resolution, frame shape and a sound switch, plus an Enhance Prompt button that turns a rough note into a fuller direction.

Use it when the clip has to fit between material you already have. The opening can match the shot that comes before it in your edit, and the closing image can be a still your client already approved. That leaves only the motion in the middle to generate. For the same reason, pick the frame shape to fit your two images before you run, not after.

Explore more first to last frame

First to Last Frame Use in Reference To Video Kling 3.0 Kling O1 Pro Seedance 2.5 Kling 4.0 Wan 3.0 Prime MiniMax H3 Max
Capability Snapshot

Inputs, Settings and Output at a Glance

What goes in, what you can set and what you get back from each run.

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 2,048 characters

Video Duration

4 to 15 seconds

Resolution

480P or 720P

Aspect Ratio

Auto, 21:9, 16:9, 4:3, 1:1, 3:4, 9:16

Audio

Switch on or off, on by default

Output

1 video per run

Getting More From Your Two Frames

Let the First Frame Set the Shape of the Clip

Before you upload, crop both images to the same shape, and make it the shape you want the finished clip to have. ByteDance's video generation documentation says that with an adaptive ratio, first-and-last-frame jobs use the ratio closest to the uploaded first frame image, so the opening image decides the framing. In practice: if you switch Aspect Ratio to Auto, compose the First Frame at the exact shape you want. If you pick a fixed ratio instead, crop both frames to it so the Last Frame isn't the odd one out when the move lands.

How Do You Make a Video From a First and Last Frame?

Six steps take you from two stills to a finished clip.

1

Add the opening image in "First Frame"

Click "Upload first frame" in "First Frame" and choose the image the clip should open on. This upload is required. Treat it as the first shot of the scene, because the motion starts from its framing, subject placement and lighting. The accepted file types and size limits are in the FAQ below.

2

Add the closing image in "Last Frame"

Use "Upload last frame" in "Last Frame" for the image the clip should end on. This one is required too. Keep the subject, setting and light consistent with the first frame, and change only what the shot is about, such as the box opening or the lights coming on. That gives the generator one clear change to animate instead of a jump between two unrelated scenes.

3

Describe the move in the "Prompt" box

"Prompt" is required and takes up to 2,048 characters. Spend them on what happens between the two images: the action, the camera movement, the pace and any sound you want. There's no need to re-describe what the images already show. If you only have a rough idea, the "Enhance Prompt" button under the box suggests fuller versions, and your text doesn't change until you pick one.

4

Set "Video Duration", "Resolution" and "Aspect Ratio"

"Video Duration" runs from 4 to 15 seconds in one-second steps and starts at 4. Give bigger changes between the frames more seconds so the move doesn't feel rushed. "Resolution" offers 480P (the default) and 720P: test your prompt at 480P, then switch to 720P for the take you'll keep. "Aspect Ratio" offers Auto, 21:9, 16:9, 4:3, 1:1, 3:4 and 9:16 and starts on 1:1, so change it to match your frames. Use 21:9 for cinematic widescreen and 9:16 for vertical social clips.

5

Check "Generate Audio", then click "Generate"

Before you click "Generate", check the "Generate Audio" switch. It's on by default, so the clip comes with sound unless you turn it off. Leave it on if your prompt describes ambience or sound effects. Turn it off if the clip will get music or a voiceover later in your edit.

6

Review and download the result

Play the clip next to your two images. Check that it opens on your first frame, that the move reads as one continuous action, and that it settles on your closing composition. If it does, download it. If not, adjust the prompt or the duration and run it again before you start replacing the images.

Which Seedance 2.0 Mini Workflow Fits Your Shot?

How the two-frame workflow compares with the other ways to start a Seedance 2.0 Mini clip.

What you need before the first run You need a First Frame image, a Last Frame image and a prompt, so you start with two finished stills in hand. Text To Video builds the clip from a written prompt, so there are no images to prepare beforehand. Pick the two-frame workflow when the opening and closing images already exist, such as approved packshots or storyboard keyframes. Pick Text To Video when you're still exploring the idea and have nothing to upload.
How the end of the shot is decided You upload the closing image yourself, so you decide where the shot ends before the run starts. Image To Video, the first-frame workflow, takes a single image and uses it to open the clip. Choose the two-frame workflow when the last moment has to match a specific picture, such as a logo lockup or a finished product shot. Choose Image To Video when only the opening needs to come from an image.
Use first-to-last-frame when both ends of the shot already exist as images. If the ending is still open, start from one image or from text instead.

Fixing Common Two-Frame Problems

Problems you're likely to hit with two-frame runs, and what to change before you try again.

My frame image won't upload.

Each frame box only accepts certain image types, with a maximum file size and a minimum pixel size. The exact limits are in the FAQ on uploads. Re-export the image in an accepted format, compress or resize it if the file is too heavy, or use a larger original if it's too small. Then upload it again. Nothing else on the form needs to change.

The Prompt box won't take my whole description.

The Prompt box stops at its character limit, which is listed in the prompt step above. Long prompts often spend most of their length describing what the two images already show. Cut appearance details like colours, wardrobe and set dressing. Keep the action, the camera movement and the timing, since neither image can carry those. Then run it again with the shorter prompt.

My widescreen or vertical frames came out square.

Aspect Ratio starts on a square default. If you didn't change it, the run uses that shape, not the shape of your images. Set Aspect Ratio to match your frames: the wide option for panoramic stills, the tall one for vertical shots. Crop both images to that shape before uploading so neither has to be refitted, then generate again.

I need a longer transition than one clip allows.

A single run can't go past the longest Video Duration option, so split a long change into stages. Generate the first stage. For the next run, put that stage's Last Frame image into First Frame and upload a new closing image into Last Frame. Each join then has the same image on both sides of the cut, which makes the segments much easier to line up in your edit.

Frequently Asked Questions

Do I need both a first frame and a last frame?

Yes. First Frame and Last Frame are both required uploads, so a run needs two images: one for how the clip opens and one for how it closes. If you only have an opening image, Image To Video is the workflow for a single starting picture.

Do I still have to write a prompt if both frames are uploaded?

Yes, the prompt is required in this workflow too. Use it for what the images can't show: what moves, how the camera travels, how fast the change happens and what it should sound like. A rough idea is enough to start with. Enhance Prompt can suggest a fuller version, and you decide whether to use it.

Which image files can I upload as the first and last frame?

Each frame takes one image in JPG, JPEG, PNG or WEBP format. Files can be up to 10 MB and must be at least 300×300 pixels. A run uses exactly two images, one in First Frame and one in Last Frame. Before you upload, crop both to the shape you plan to pick in Aspect Ratio.

How many videos do I get from one run?

One. Each generation returns a single clip. To compare versions, such as a different camera move or a longer duration, change the prompt or the settings and run it again with the same two images.

What makes one generation cost more than another?

The credit cost of each run is calculated from the options you choose, and it's shown above the Generate button once you're signed in. It changes with Video Duration and Resolution. Using Enhance Prompt for suggestions doesn't use credits.