Start from words
A written description is enough. Name the subject, the action and the camera, and Seedance 2.0 builds the shot — lighting, motion and lens behaviour included. Best when nothing exists yet and you are still finding the idea.
AI Video Generator
AI video generation
AI video generation
Describe the scene, the character action, and the camera move. e.g. “slow dolly-in on a lone figure crossing a neon-lit street in the rain”.
Model
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
- - AI video generation
- - Describe the scene, the character action, and the camera move. e.g. “slow dolly-in on a lone figure crossing a neon-lit street in the rain”.
- - Generation time varies with the model, clip settings, and provider load.
- - Use up to 9 reference images to guide a character, product, or look; generated details may still vary.
- - 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.
AI video generator — turn a sentence into a finished clip with sound
This is the general-purpose entry point to FrameAI: one generator, powered by ByteDance’s Seedance 2.0, that covers every route into a video. Start from a written idea, a photograph, a clip you already shot, or any combination of the three, and get back a 4-to-15-second film at up to 4K with audio generated in step with the picture.
FrameAI supports 480p through 4K, synchronized audio, seven aspect ratios, and commercial use for paid-plan output under the Terms of Service. Credit cost is quoted from your settings before you generate, and a failed job refunds itself.
- One generator, every input type
- Up to 4K with synchronised audio
- Commercial use on paid plans
Three ways into the same generator
You do not pick a product. You pick what you happen to have, and the model works from there.
Start from a picture
Upload a photo, a product shot, an illustration or a frame, and describe what should happen. The image anchors the look while the prompt supplies the movement — the fastest way to get a specific real object into motion.
Start from footage
Feed in up to three video clips as references and the model borrows their motion, pacing or style for a scene they never contained. Useful when the move matters more than the material.
How AI video generation actually works, and what that means for your prompt
A video model does not assemble frames one after another the way an editor does. It generates the whole clip as a single object, denoising it from a compressed representation while every frame is conditioned on every other frame plus on whatever you supplied — your text, your images, your reference footage. That is why temporal consistency is a property of the model rather than something you fix in post, and why a subject can stay recognisably itself while the camera moves around it.
The practical lesson is that prompts should describe a shot, not a story. The model has four to fifteen seconds; it can execute one clear action with one clear camera behaviour beautifully, and it will make a mess of five plot beats crammed into a paragraph. Write like a director filling in a shot list — subject, action, camera, light, mood — and reach for the extend and transition tools when a sequence genuinely needs to be longer than one take.
Describe one shot
"Slow dolly-in on a barista pouring latte art, warm window light, shallow depth of field" beats a paragraph of narrative. Concrete nouns and camera verbs do the work.
Change one thing at a time
When a result is close but wrong, edit a single variable and regenerate. Rewriting everything teaches you nothing about which word actually mattered.
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 an AI video in three steps
From blank page to downloadable clip, without opening an editing suite.
- 1
Describe the shot
Type what should be on screen, in English or Chinese. Attach reference images or clips if you have them — they are optional, and each one gives the model something more to hold constant.
- 2
Choose your output
Set the aspect ratio for wherever this is going, a duration between 4 and 15 seconds, and a resolution from 480p to 4K. The credit cost updates as you choose, before anything is spent.
- 3
Generate and download
The job runs in the background and lands in My Creations. If it was generated on a paid plan, you may use it commercially under the Terms of Service; if generation fails, the credits return automatically.
What an AI video generator is genuinely good for
The wins are where volume, speed or cost used to make video impossible — not where a film crew was already the right answer.
Marketing that needs ten versions
Ad testing, seasonal refreshes and localised cuts all need the same idea many times over. Generating variants costs minutes instead of a reshoot, and every one arrives in the right aspect ratio.
Product and e-commerce video
A single product photo becomes a rotating hero shot, a lifestyle scene or a detail pass. Reference images keep the actual product accurate rather than approximated.
Concepts, pitches and prototypes
Showing a client the film beats describing it. Generating pre-visualisation at 4K makes an idea concrete before any real budget is committed to it.
AI video generation — frequently asked questions
What is an AI video generator?
It is a model that produces video directly from your inputs — text, images, reference footage — rather than editing existing clips together. FrameAI runs ByteDance’s Seedance 2.0, which generates 4-to-15-second clips at up to 4K with synchronised audio.
Do I need video editing experience?
No. There is no timeline, no keyframing and no plugin chain. You write a description, choose a ratio and length, and press generate. The skill that helps is describing a shot precisely, which is learned in a handful of attempts.
How long can the generated videos be?
Each generation produces 4 to 15 seconds at 24 fps. For longer pieces, use Video Extend to continue a clip or Video Transition to join two takes — building a sequence from good shots works far better than asking for one very long one.
Does the generated video include sound?
Yes. Seedance 2.0 generates sound effects, ambience and score in step with the picture, so footsteps and impacts land on the frame where they belong. You can turn audio generation off if you plan to score the clip yourself.
How much does it cost?
You spend credits, and the amount depends on duration, resolution and aspect ratio. The exact cost appears in the generator before you commit. New accounts start with free credits, and any generation that fails on the provider side is refunded automatically.
Can I use the videos commercially?
Yes. If the video was generated while you had an active paid plan, you own it and may use it commercially under the Terms of Service. Output generated with free credits is for evaluation only.
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