An ending you can rely on
When a shot must finish on a product, a logo or a specific composition, hoping is not a plan. A pinned final frame turns the ending from an outcome into a requirement.
First & Last Frames
Endpoint-controlled AI video
Supply an opening image and a required closing image, then describe the motion between them. Generate AI video with pinned endpoints — perfect for loops, product reveals, and seamless edits.
Pin the opening frame; add a closing frame when this workflow requires one — the model fills the motion between them.
Model
Input modeFirst / Last frame
Pin the opening frame; add a closing frame when this workflow requires one — the model fills the motion between them.
0/2000
4s15s
Outputs
Audio
With audio
Synchronized sound, voice and music
Credits required:36
Your generated video appears here. Describe a shot and hit Generate.
My creationsDownload
Generation tips
- - Supply an opening image and a required closing image, then describe the motion between them. Generate AI video with pinned endpoints — perfect for loops, product reveals, and seamless edits.
- - Pin the opening frame; add a closing frame when this workflow requires one — the model fills the motion between them.
- - Generation time varies with the model, clip settings, and provider load.
- - Name the subject, the action, and the camera move. Vague prompts give vague shots.
- - Longer clips cost more credits — start at 5s to test a look, then extend.
- - Audio is generated with the picture; describe the sound you want in the prompt.
First and last frames — guide both ends of a generated shot
Upload a required opening image and closing image, then describe the motion between them. The provider generates a new clip using both endpoints as input. The endpoints guide the result, but the middle can still drift and should be reviewed.
This workflow is useful when a shot needs a planned beginning and ending: a product reveal, a handoff to another scene, or a loop concept. The final frame is an input rather than a promise that the output will match every pixel; choose compatible images and keep the prompt focused on the path between them.
- Required first and last images
- Prompt describes the path between them
- Review the generated middle
What pinning both ends gives you
One constraint, three consequences that matter when footage has to fit somewhere specific.
Seamless loops
Use the same image as both the first and last frame and the clip returns exactly to where it started — the cleanest way to produce a background loop with no visible restart.
Shots that hand off cleanly
Ending one shot on the frame the next one begins with makes an edit continuous. This is how you assemble a sequence that does not announce every cut.
Choosing two frames the model can actually connect
The distance between your two frames sets the difficulty. Images that share a subject, a viewpoint and a lighting scheme describe a short, plausible journey, and the model fills it convincingly. Two images with nothing in common describe a journey that has to be invented wholesale, and the middle of the clip is where that invention becomes visible. If the endpoints are genuinely far apart, either give the shot more time or reconsider whether this should be one shot at all.
Duration is the other half of the equation, and it is the one people get wrong. The same pair of frames over four seconds produces a fast, direct movement; over fifteen it produces a slow one with a great deal of invented material in between — which is more room for drift, not less. Match the length to how far apart the frames are, describe the path you want taken rather than just supplying the endpoints, and keep both images at similar resolution and lighting so the clip does not appear to change exposure as it plays.
Related frames, short journey
Shared subject, similar viewpoint, consistent lighting. The closer the two ends, the more convincing the middle.
Match duration to distance
Nearby frames over a long duration leave the model inventing filler. Distant frames over a short duration produce a rushed morph.
Model specifications
What Seedance 2.0 actually delivers on FrameAI
Published limits, not marketing rounding. Every number below is enforced by the generator before your credits are reserved, so a job that would exceed a limit is rejected instead of failing halfway.
- Up to 4K
- Output resolutions — 480p / 720p / 1080p / 4K
- 4–15 sec
- Clip length per generation, at 24 fps
- Native audio
- Sound effects and ambience generated in sync
- 9 + 3
- Reference images plus reference videos, one request
- 7 ratios
- Supported aspect ratios — Auto, 16:9, 4:3, 1:1, 3:4, 9:16, 21:9
- Commercial use on paid plans
- Subject to the Terms of Service
Why creators run Seedance 2.0 on FrameAI
The model is ByteDance’s. What FrameAI adds is the part that decides whether it is usable at work: honest limits, honest billing, and output you are allowed to ship.
Every input in a single pass
Text, up to nine reference images and up to three reference video clips are fused in one generation, not stitched afterwards. There is no separate text-to-video mode to switch into — you add whatever material you have and the model reconciles it.
Sound generated with the picture
Seedance 2.0 writes footsteps, room tone, impacts and score alongside the frames, locked to the action. Most models hand you a silent clip and leave the sound design to you; here the export is already finished.
Seven ratios, up to 4K
One prompt covers a 9:16 vertical cut for Reels, a 16:9 master for YouTube and a 21:9 anamorphic frame for a title sequence. Resolution runs from 480p up to 4K, so the same take can serve a cinema-shaped frame or a phone.
Credits you can predict
The exact credit cost is calculated from your duration, resolution and ratio and shown before you press generate — never after. If a generation fails on the provider side, the reserved credits are returned automatically.
Built for iteration, not for queuing
Seedance 2.0 Fast and Seedance 2.0 Mini exist for the twenty drafts nobody sees. Rough a shot at 480p for a fraction of the credits, then re-run the take that worked at full resolution on the flagship model.
Yours to sell
Paid-plan output may be used commercially under the Terms of Service. Free-credit output is for evaluation only.
Author a pinned shot in three steps
Two images and a sentence about the path between them.
- 1
Choose your opening frame
The composition the shot starts on. Sharp, well-lit and framed with a little room for the camera to move.
- 2
Choose the frame it must land on
The ending the edit requires. For a perfect loop, use the opening image again here.
- 3
Describe the path and generate
Say how the shot travels between the two — the camera move, what the subject does — and set a duration proportional to the distance.
Where pinned endings are non-negotiable
Each of these has a requirement that a text prompt alone cannot guarantee.
Ads that must end on the pack shot
Brand guidelines usually specify the final frame precisely. Pinning it makes compliance structural rather than a matter of regenerating until you get lucky.
Looping backgrounds
Hero sections and ambient displays need motion with no visible restart. Identical first and last frames produce a genuinely seamless loop.
Sequences built to cut together
When each shot ends on the frame the next begins with, the edit flows continuously — which is difficult to achieve any other way with generated footage.
First and last frames — frequently asked questions
What is first and last frame generation?
You supply a required opening image and closing image, then the provider generates a new clip between them from your prompt. The two images guide the endpoints; the generated middle can still vary.
Can I make a perfectly looping video?
You can try using the same image at both ends to guide a loop, but the generated middle and the actual join still need review. It is not a guarantee of a perfect loop.
What if my two frames are very different?
The model has to invent a long path between them, and that invention shows up in the middle of the clip. Either choose endpoints that share a subject, viewpoint or lighting scheme, or split the idea into two shots joined by a transition.
How do I choose the duration?
Match it to how far apart the frames are. Nearby frames over fifteen seconds leave the model inventing filler; distant frames over four seconds produce a rushed morph. Adjust duration before rewriting the prompt.
Can I still describe what happens in between?
Yes, and you should. The frames set the endpoints; the prompt sets the path — the camera move and the subject action that get you from one to the other.
How is this different from video transition?
The mechanism is similar, the intent is not. First and last frames authors a single shot whose ends you control. Video transition connects two shots that already exist in an edit.
Keep going with the rest of the toolkit
Each tool is the same Seedance 2.0 model pointed at a different job. Move between them freely — your credits, history and exports are shared.
Generate your first Seedance 2.0 video now
Write one line, pick a ratio, press generate. No editing suite, no render farm, no post-production pass — a finished clip with sound, ready to download.
- Free credits on sign-up
- Paid commercial use
- Refunded if a generation fails