Native Audio, Generated Jointly
Sound is produced together with the picture rather than dubbed on afterwards, so effects and ambience land on the right frame. The clip above is raw Seedance 2 Mini output at 720p.
La Tormenta de Arena AI2 photos · 15s · from 121 credits
Tip Toeing in My Jordans AI1 photo (+1 optional) · 10s · from 81 credits
Hotel Lobby AI2 photos · 10s · from 81 credits
Wipe Me Down Dance AI1 photo · 13s · from 105 credits
Bulin 47 AI1 photo · 10s · from 81 credits
AI NEON Dance2 photos · 23s · from 185 creditsSeedance 2 Mini is the compact member of ByteDance's Seedance 2 family. It generates video with synchronized native audio from a prompt, a first and last frame, or a set of reference images and audio clips — starting at 24 credits per second.
From prompt to finished clip
Choose Seedance 2 Mini from the Viberot video model selector. No setup or API keys required.
Write the scene, subject, camera movement and the sound you want. Prompts can run to 20,000 characters, so there is room for detailed direction.
Go text-to-video, pin a first and/or last frame, or switch to reference mode for up to nine images. These modes are mutually exclusive — pick the one that fits the shot.
Choose 480p or 720p and any duration from 4 to 15 seconds, then generate. The finished clip lands in your library, watermark-free on paid plans.
Seedance 2 Mini follows the same flow as every other video model on Viberot — pick the model, describe the shot, choose your inputs, and generate.
Seedance 2 Mini keeps the parts of Seedance 2 that matter most for everyday work — joint video and audio generation, precise frame control, and rich reference input — while cutting the per-second cost to a fraction of the flagship.
Sound is produced together with the picture rather than dubbed on afterwards, so effects and ambience land on the right frame. The clip above is raw Seedance 2 Mini output at 720p.
Pin the opening frame, the closing frame, or both — each slot uploads independently, so you can define an ending without supplying a start.
Multimodal reference mode accepts up to nine images plus reference audio, so characters, props and sound design stay consistent across a sequence.
480p costs 24 credits per second and 720p costs 51. A 5-second 480p draft runs 120 credits — cheap enough to iterate on ideas before committing to a flagship render.
Seedance 2 Mini is the model to reach for when you are still exploring. Block out the idea here at 480p, then re-render the shot you like on Seedance 2.0 or Kling 3.0 once the direction is locked.
Seedance 2 Mini sits between the ultra-cheap draft models and the flagship renderers. Here is where it lands against the rest of the Viberot lineup.
Unlike HappyHorse 1.1, Kling 3.0 Turbo or Grok Imagine 1.5, Mini generates a real audio track alongside the video.
Roughly half the per-second cost of Seedance 2.0 at the same resolutions, with the same input modes.
Any duration on a slider, billed per second — no jumping between fixed 5s and 10s tiers.
Matches the flagship on reference capacity, plus up to three reference audio clips.
Upload a closing frame without a starting frame — useful for reverse-engineering a shot toward a known ending.
Caps at 720p. For 1080p or 4K delivery, step up to Seedance 1.5 Pro, HappyHorse 1.1 or Kling 3.0.
Common questions
Native audio, nine reference images and frame-level control from 24 credits per second. Create a free account and generate your first clip in under a minute.