Seedance 2.0

About

ByteDance's Seedance 2.0 generates high-quality video from text prompts and multimodal references, with fixed 24 fps output and durations up to 15 seconds.

Full Seedance 2.0 supports T2V, I2V, image+audio-to-video, and V2V workflows, including 1080p and native 4K paths. Mini is the faster, lower-cost 720p tier for iteration. It replaces Seedance 2.0 Fast; legacy Fast model IDs are automatically routed to Mini.

Variants

Seedance 2.0 ships in 2 builds — same family, different speed/quality trade-offs. Pick the variant that fits, or start with the default.

Photos and clips of real people

Seedance 2.0 and Mini accept your own likeness, or that of people who gave you explicit, legal consent, once your account accepts a one-time likeness agreement in the Sogni app. Without it, they refuse a photo or video that may contain a real person. It is the one agreement for Seedance video generation models, so if you already accepted it, you are covered. For mature themes, use Seedance 2.0 Mini Uncensored below.

• Best likeness. Likeness holds best in Reference to Video, with the person's photo attached as a reference, for example Use @Image1 as the presenter. Image to Video, with the photo as the first frame, can drift away from the face on big head turns.
• API and SDKs. The agreement belongs to your account, so once you accept it in the Sogni app, seedance-2-0 and seedance-2-0-mini jobs through the API and SDKs accept real people too.

Seedance 2.0 Mini Uncensored

Seedance 2.0 Mini Uncensored (model ID seedance-2-0-mini-uncensored) is the Mini tier on this page with every capability and limit unchanged: 4–15 second clips at 480p or 720p with native audio, and up to 9 images, 3 videos and 3 audio references. It runs on a more permissive content policy for mature themes. Full Seedance 2.0 has no uncensored version; for 1080p, clips up to 30 seconds, or Edit and Extend, use Seedance 2.5 Uncensored.

• Agreement. Before the first render, every account accepts the one-time likeness and consent agreement in the Sogni app; the same agreement covers all Seedance video generation models.
• Mature themes. Seedance needs detailed reference images or videos that show the anatomy and actions you want. Wan 3 Uncensored needs far less guidance and usually just works.
• Where to find it. At app.sogni.ai it appears with the other Seedance models in the Seedance workflows while the Sensitive Content Filter is off (18+). Through the API and SDKs use seedance-2-0-mini-uncensored, or seedance2-mini-uncensored in Creative Agent workflows and the Creative Agent Skill. Accept the agreement once in the Sogni app first; until then, jobs are refused with error 4103 and nothing is charged.

Seedance 2.0 Mini Uncensored pricing

Seedance 2.0 Mini Uncensored comes at a premium rate above standard Seedance 2.0 Mini. It is paid with Premium Spark on every plan, with 5% off for Unlimited and 10% off for Unlimited Pro.

Without a reference video. With one, the reference seconds and output seconds bill at one lower per-second rate: a 15-second reference plus 15 seconds of 720p output bills as 30 seconds, about 680 Spark ($3.40), versus 567 Spark ($2.84) for 15 seconds without one.

Pricing

Seedance 2.0 is a pay-as-you-go Premium Spark vendor model. Every clip uses Premium Spark (bought or awarded, never Free Spark), including for Sogni Unlimited members. Unlimited saves 5%; Unlimited Pro saves 10%.

Rows marked no video reference cover output only; video-reference rows include the input and output durations shown. With a video reference, each clip charges for input-video seconds plus generated output seconds at this model’s with-video rate. Add all reference-video durations, including fractional seconds; reusing the same reference for another clip adds the input charge again.

With a reference video, the reference seconds bill at the same per-second rate as the output seconds, since reading a reference takes about as much GPU time as generating the same length. That rate is lower than the rate without a reference, but the job costs more in total: on Seedance 2.0, a 15-second reference plus 15 seconds of 720p output bills as 30 seconds, about 724 Spark ($3.62), versus 590 Spark ($2.95) for 15 seconds without one. Input counts as at least two-thirds of the output duration, rounded up to a whole second. The quote before rendering assumes a 15-second reference; the final charge uses the actual length.

API

One Sogni API key reaches every model in the catalog — call Seedance 2.0 with the exact model id for the variant you want.

import { SogniClient } from '@sogni-ai/sogni-client';

const client = await SogniClient.createInstance({
  appId: crypto.randomUUID(),
  apiKey: process.env.SOGNI_API_KEY,
  network: 'fast',
});

const project = await client.projects.create({
  type: 'video',
  modelId: 'seedance-2-0',
  positivePrompt: 'a slothicorn surfing a wave of liquid paint, slow push-in, cinematic',
  numberOfMedia: 1,
  duration: 5,
});

const [url] = await project.waitForCompletion();
console.log(url); // result link — download within 24h

curl https://api.sogni.ai/v1/creative-agent/workflows \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $SOGNI_API_KEY" \
  -d '{
    "input": {
      "title": "First render",
      "steps": [{
        "id": "step1",
        "toolName": "generate_video",
        "arguments": { "prompt": "a slothicorn surfing a wave of liquid paint, slow push-in, cinematic", "videoModel": "seedance2", "duration": 5 }
      }]
    },
    "confirm_cost": true
  }'

Example uses seedance-2-0. Use any exact model id from the Variants table above. Full reference at docs.sogni.ai.