Describe the next beat
Say what should happen after the source clip — the action, subject behaviour, and camera language. This gives the provider a useful direction without claiming that it will copy the original motion.
Video Extend
AI video extension
Upload a source clip and describe a continuation — FrameAI generates a new AI video clip that extends the scene's subject, style, and momentum. Ideal for building longer sequences.
MP4 or MOV, 2–15s each, 15s total, up to 200MB. Requires R2 with a public domain — the model fetches the file over the network.
Model
Input modeReferences
Reference videos 0/3
MP4 or MOV, 2–15s each, 15s total, up to 200MB. Requires R2 with a public domain — the model fetches the file over the network.
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
- - Upload a source clip and describe a continuation — FrameAI generates a new AI video clip that extends the scene's subject, style, and momentum. Ideal for building longer sequences.
- - MP4 or MOV, 2–15s each, 15s total, up to 200MB. Requires R2 with a public domain — the model fetches the file over the network.
- - 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.
Video continuation from a source clip
This page is a guided reference-video workflow: upload a source clip and describe the next beat. The provider generates a new 4-to-15-second clip inspired by that reference; it does not perform frame-perfect native extension.
The source video is sent as reference input, not as an editable timeline or a guaranteed final-frame condition. Subject, style, motion, and fine detail can drift. Use a clear source ending, review each result, and only then decide whether to run another separate generation at the selected model, resolution, ratio, and duration.
- Source video as reference
- Newly generated continuation
- Review each step for drift
What to expect from source-video continuation
A focused source and a specific next action improve the result, but the new clip is still a separate generation.
Keep the source focused
A short reference with a clear subject and readable ending is easier to use than a long or heavily blurred clip. The provider accepts up to three reference videos within the configured total-duration limit.
Review before chaining
Each continuation is a new task. Check identity, lighting, motion, and the cut before using the result as inspiration for another generation.
How to get a more useful continuation
Choose a source whose subject and final moment are easy to read. A stable ending can give the provider better visual context, but it is not a promise of a seamless join because the workflow uses reference-video generation rather than native frame extension.
Describe what happens next instead of pretending the original timeline is being edited: "she turns toward the window and the camera settles" is more useful than a claim that the exact motion will be preserved. Repeat the lighting, palette, and wardrobe details that matter, then inspect the new clip for drift.
Use a readable ending
A visible subject and a stable final moment provide clearer reference material. They improve the odds of a useful result but do not guarantee a seamless continuation.
Generate in reviewable steps
Shorter separate generations give you a checkpoint. Keep the outputs that work and revise the prompt when the subject, motion, or style drifts.
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.
Generate a continuation in three steps
Upload a source, describe the next beat, and review the new clip before doing anything else.
- 1
Choose the source clip
Upload a focused reference video that ends on a clear moment. The current workflow uses the clip as reference material, not as a frame-accurate edit source.
- 2
Describe what happens next
Write the new action and camera behaviour, plus the continuity details that should guide the model. Avoid promising that the original motion will be copied exactly.
- 3
Generate and review
Inspect the result at the cut. If it is useful, you can start another separate generation from a suitable reference; if it drifted, adjust the prompt and try again.
When reference-based continuation helps
Use this workflow when you have a source clip whose subject, look, or final moment should guide a newly generated follow-up.
Give a hero shot another beat
When a source clip has the right subject and look but needs a new moment, use it as reference and describe the action you want to explore next.
Explore a performance variation
Use the source as visual guidance for a reaction, gesture, or reveal, while expecting the generated performance and fine details to vary.
Create options for an edit
Generate separate follow-up options and choose the one that fits your edit. Do not treat the page as a guarantee of an exact-duration or frame-perfect continuation.
Video extend — frequently asked questions
What is video extend?
It is a guided workflow that accepts a source video as reference and generates a new clip based on a prompt for what happens next. It is not a frame-perfect native extension tool.
How long can I extend a clip?
Each extension is itself a generation of 4 to 15 seconds, and extensions can be chained. Shorter steps are recommended — they give you a checkpoint before drift accumulates.
Will the extension match the original exactly?
No exact match is guaranteed. The new clip is generated from the source-video reference, so identity, motion, lighting, audio, and fine details can change. A clear source ending may provide better guidance.
Can I extend a video I did not generate here?
Yes — upload it as a reference clip. Reference video is capped at three clips totalling fifteen seconds, so trim to the ending segment you actually want continued.
Why is there a visible seam?
A visible difference is expected when the provider interprets the source as reference material. Motion blur, fast camera movement, occlusion, or an unclear subject can make the change more noticeable; try a clearer source and a more specific prompt.
Does the extension include audio?
Audio can be generated per task when the model setting is enabled, but continuity with the source audio is not guaranteed. Describe the intended ambience and review the result before editing it into the source.
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