We use cookies to enhance your experience, provide personalised content, and analyse our traffic; see our Cookie Policy.
bytedance/seedance-2Short, punchy shots with references and sound
Seedance 2.0 is ByteDance's video model, launched on 12 February 2026, with an architecture that generates picture and sound together. It produces up to 15 seconds of audio-video at 480p or 720p, with multiple shots possible within one clip.
Its strength is control by reference. ByteDance says it takes up to 9 images, 3 videos and 3 audio clips as input to pin a character's look, a style or a movement. Ideal for ad creatives, product demos, social formats and storyboard animatics.
On Tchavi you pay with mobile money, no bank card and no KYC. One account and one API key open 80+ models, video, image and text together, and your credits never expire.
Set the duration, or let the model choose (duration: -1)
Dialogue, sound effects and music synced to the scene
Pin a character's look or a style
Animate a photo or lock the final frame
16:9, 9:16, 1:1, 4:3, 3:4 or adaptive to your references
Set a seed to get a similar render back
A bottle of bissap juice on a counter, the camera orbits it, condensation drops slide down, ice cubes clink, market ambience behind. An 8-second ad shot.
Shot 1: a young designer adjusts a wax-print fabric on a mannequin. Shot 2: she smiles at the camera and says "It's ready!" Natural workshop light, sewing-machine noise.
Seedance 2 Fast (model: "seedance-2-fast") is the draft tier: it trades quality for speed and cost, to find a shot. Move to Seedance 2 once the concept is settled and you want the final render.
Seedance 2 takes reference images, videos and audio to keep a character consistent; Kling V3 (model: "kling-v3") goes up to 4K and accepts a negative prompt. Need references: Seedance; need high definition: Kling.
Duration: up to 15 seconds, or -1 to let the model choose. Resolutions: 480p or 720p. Formats: 16:9, 9:16, 1:1, 4:3, 3:4 or adaptive. Audio is on by default; generate_audio: false turns it off.
According to ByteDance, the platform takes up to 9 images, 3 videos and 3 audio clips as references, along with the prompt text. Name each reference in the prompt to say what it contributes: the face, the setting, the rhythm.
Send a POST to /v1/videos/generations with your Tchavi key and model: "seedance-2", then prompt, duration, resolution ("480p" or "720p") and aspect_ratio. Add image_url to animate a photo, or references to pin a character.
Yes. Top up your Tchavi account with Orange Money, MTN MoMo, M-Pesa or another mobile money operator. No bank card is needed. Credits land as soon as the payment is confirmed and are deducted as you use them.
Yes. Keep your current OpenAI SDK, just change the base URL to Tchavi's (shown in the API tab) and use your Tchavi key. The rest of your code stays the same.
import Tchavi from '@tchavi/sdk';
const client = new Tchavi({ apiKey: 'API_KEY' });
const job = await client.videos.generations.create({
"model": "seedance-2",
"prompt": "A cinematic shot of a cat astronaut floating in space",
"resolution": "480p",
"duration": 5,
"aspect_ratio": "16:9",
"generate_audio": true
});
console.log('Job id:', job.id);
console.log('Credits used:', job.tchavi.credits_used);
// Poll for completion
let result = await client.videos.generations.retrieve(job.id);
while (result.status === 'pending' || result.status === 'processing') {
await new Promise((r) => setTimeout(r, 2000));
result = await client.videos.generations.retrieve(job.id);
}
console.log('Video URL:', result.output?.video_url);Every parameter this model accepts on the API. Check the SDK for full typings.
modelRequiredID of the video model (e.g. `seedance-2`).
promptRequiredText description of the video to generate.