Seedance 2 Mini is the entry tier of the Seedance 2 family: 480p or 720p video from text, an image or a source clip, billed per second. It is the tier to use when you need several attempts rather than one expensive one.
About half the per-second rate of Seedance 2.0 at the same resolution.
Feeding it video costs less per second than generating motion from nothing.
AI-generated reference clips picked to show the spread of work a cheap tier gets used for — fast real-world motion, stylised animation, and a sequence that cuts between shots. They illustrate the territory rather than being outputs captured from this specific model, and every generation is different anyway.
Seedance 2 Mini is the lowest tier of ByteDance's Seedance 2 family, sitting below Seedance 2.0 and Seedance 2 Fast. It takes the same three kinds of input the family does — a written description, a still image, or an existing clip to work from — and caps out at 720p rather than 1080p. What it is really for is run count.
Video generation is unpredictable in a way image generation is not: a prompt that reads perfectly can produce motion that drifts, a face that changes, or a camera move that never happens, and there is no way to know except to run it. At the top tier each of those failures is expensive enough to make you cautious, which is the wrong instinct when the fix is another attempt.
Mini makes attempts cheap enough that you can afford to be wrong. Billing is per second of output, and the rate drops further when you supply a source video rather than asking for motion from a still.
Creative engine
Rate is per second and depends on resolution and whether you supplied a source video.
Below Seedance 2 Fast and Seedance 2.0, and capped at 720p instead of 1080p.
Text to video, image to video, or working from a source clip you supply.
Cheap enough per run that a failed take is a retry rather than a decision.
Pick Mini when you do not yet know whether the shot works. That covers more of the process than people expect: finding the prompt, testing whether a camera move survives the render, checking that a character holds together for the length you need, and generating filler or background footage where nobody will be examining the frame. Move up to Seedance 2 Fast or Seedance 2.0 for the version you actually ship, once the prompt is settled and 1080p is worth paying for. The workflow that wastes the least money is to iterate down here and finish up there — the prompt you arrive at transfers between tiers.
Settle the wording, the motion and the framing at the low tier, then run the final higher up.
Working from existing video costs less per second than generating the motion from a still.
Billing is per second, so testing on a short clip is the cheapest way to learn anything.
What the entry tier covers, and where the tiers above it pick up.
Generate a clip from a written description with no source material.
Start from a still and describe the motion you want added to it.
Work from a clip you supply, at a lower per-second rate than generating motion outright.
Two resolutions, with 1080p available on the tiers above.
The generation parameters for this tier, based on the KIE (kie.ai) spec it runs on here.
Four steps, and the last one is the one people skip.
Text, a still image, or a source clip — the last of these is also the cheapest per second.
State the camera move and what moves in frame. Adjectives survive a render far less reliably than stated actions do.
480p and a short duration while you are testing; both are what the cost is made of.
Video output varies run to run. Two or three attempts at this tier still cost less than one at the top.
Jobs where run count matters more than the last increment of resolution.
Find out whether the motion you described is a thing the model can actually do.
Short vertical video where 720p is the delivery resolution anyway.
Footage that sits behind text or plays under a cut, where nobody inspects the frame.
Restyling or extending a clip you already have, at the lower video-input rate.
One credit pool covers Nano Banana images and Seedance and Veo video. The cost shows before every run, and failed jobs are not charged. Use a subscription for ongoing work, or a one-time pack when you just need to top up.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
One-time top-ups — buy extra credits any time you run low.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
Charges appear as “KAVEL AI” on your card statement. You can cancel any time from Settings → Billing; cancellation takes effect at the end of the period you already paid for.
Powered by
Use one balance across every supported image and video model — ten of them today, listed below — and check the credit cost before each request.
Video models
Image models
Model credit guide
You're only charged for successful generations. The exact estimate in the generator varies by model, length, resolution, audio, and number of images.
| Type | Model | Credit cost |
|---|---|---|
| Video | Seedance 2.0 | 5s 720p image-to-video ≈ 188 credits; text-to-video ≈ 308 credits. Scales with resolution and length. |
| Video | Seedance 2 Fast | Faster and lower cost. 5s 720p text-to-video ≈ 248 credits. |
| Video | Seedance 2 Mini | The cheapest Seedance tier. 5s 720p ≈ 154 credits, 5s 480p ≈ 72 credits. |
| Video | Seedance 1.5 Pro | Audio doubles the rate. 5s 720p ≈ 27 credits silent, ≈ 53 with audio; 1080p ≈ 57 and ≈ 113. |
| Video | Veo 3.1 | Billed per video, not per second (Lite tier). About 45 credits at 720p and 53 at 1080p. |
| Video | Kling 3.0 | Audio raises the rate. 5s 720p ≈ 105 credits silent, ≈ 150 with audio; 1080p ≈ 135 and ≈ 203. |
| Video | MiniMax H3 | Fixed 2K, no resolution ladder. Priced per second — a 5s clip ≈ 158 credits. |
| Image | Nano Banana 2 | Generate or edit from text and images. About 8 credits per 1K image, 12 at 2K, 18 at 4K. |
| Image | Nano Banana Pro | Consistent run times across generations. About 12 credits per 1K or 2K image, 21 at 4K. |
| Image | Nano Banana 2 Lite | Faster, simpler variant. A flat 8 credits per image at every resolution. Start here to test. |
| Image | GPT Image 2 | The lowest-cost image model here. About 3 credits per 1K image, 6 at 2K, 12 at 4K. |
| Image | Seedream 5.0 Lite | Flat pricing — about 8 credits per image whether you generate or edit. |
Common questions about the entry tier of the Seedance 2 family.
Mini caps at 720p where Seedance 2.0 reaches 1080p, and it costs roughly half as much per second at the same resolution. Both take text, image or video input.
Billing is per second. A 5-second clip is about 72 credits at 480p and about 154 at 720p when generated without a source video; supplying a source clip lowers the per-second rate.
Generating motion from nothing is more work than transforming motion that already exists, and the per-second rate reflects that.
No — 480p and 720p only. For 1080p, use Seedance 2 Fast or Seedance 2.0.
Video generation is not deterministic; the same prompt produces a different take each run. That is the reason to iterate at this tier rather than a more expensive one.
Credits are returned automatically, unless the request broke the content policy.
Open the generator, describe the motion, keep it short, and run it a couple of times.
Length is the main lever on cost, so a short clip is a cheap test.
