Seedance 2.0

Seedance 2.0

The workhorse family: Standard for flagship quality and the only 4K tier, Fast for quicker drafts, Mini for the lowest cost per clip — all shooting 4–15 second takes with sound from text, images and mixed references. Live below on the same credits as every other model.

  • Failed generations auto-refunded
  • Credit packs never expire
  • Every model's price is public
  • Cancel anytime in one click
4–15 s
Single take length
4 tiers · up to 4K
480P to 4K output
9 + 3 + 3
Reference images · videos · audio
From 6 cr/s
Seedance 2.0 Mini 480P entry price

Six official takes, six full prompts

Action transfer, lip-synced dialogue, e-commerce turntables, one-line video edits and multi-clip continuations — every sample below is an official Seedance 2.0 sample paired with the complete prompt that made it. Hit Remix to load any prompt straight into the generator above and make it yours.

  • Seedance 2.0
  • Reference to video
  • 5s
  • 720P

Following the running form of the horse in 视频1, generate a golden steed galloping across a grassland, then freeze its magnificent running pose as it turns into a horse-shaped gold pendant.

  • Seedance 2.0
  • Text to video
  • 11s
  • 720P

A macro lens trained on an emerald glass frog on a leaf. The focus gradually shifts from its smooth skin to its fully transparent belly, where a bright-red heart contracts and expands with strong, steady rhythm.

  • Seedance 2.0
  • Reference to video
  • 5s
  • 720P

The two people in the picture are chatting in an office. The woman speaks first, saying: “你每次卡点到,是不是很享受这种刚刚好的感觉?” The man replies with a smile: “我有我的节奏”. As the characters speak, the conversation feels casual and natural, and matching dialogue subtitles appear at the bottom of the frame.

  • Seedance 2.0
  • Reference to video
  • 11s
  • 720P

Extract the camera from 图片1, 图片2 and 图片3, replace the background with white, and place the camera on a white table. The shot focuses on the camera in close-up, then slowly rotates around it as the subject, clearly showing the camera's front, sides and back.

  • Seedance 2.0 Video Edit
  • Video edit
  • 5s
  • 720P

Replace the perfume in the gift box in 视频1 with the face cream from 图片1, keeping the camera moves unchanged

  • Seedance 2.0
  • Reference to video
  • 12s
  • 720P

The arched window in 视频1 opens and we enter the art-gallery interior, continuing into 视频2; the camera then moves into the painting, continuing into 视频3

Two images, one clip, one soundtrack — one finished ad

ByteDance's official multimodal reference tutorial. One clip pins the first-person composition, one audio track sets the soundtrack, and two product images anchor the opening and closing frames — the prompt wires them together by number, the way material mentions work in the Molyin generator. Seedance 2.0 reads them as one brief and delivers the commercial.

Image 1 — first-frame product reference
Image 1 — first-frame product reference
Image 2 — final-frame product reference
Image 2 — final-frame product reference
Video 1 — first-person composition reference
Audio 1 — background music

The prompt

Use the first-person composition of 视频1 throughout, and use 音频1 as the background music throughout. A first-person fruit-tea commercial for the seedance-brand 「苹苹安安」 limited-edition apple fruit tea. The first frame is 图片1: your hand picks a dew-covered Aksu red apple, with a light, crisp knock of apples touching. 2–4s: fast cuts — your hand drops apple chunks into a shaker cup, adds ice and tea base, and shakes hard; the clink of ice and the shaking land right on an upbeat drum groove, background voice: 「鲜切现摇」. 4–6s: first-person close-up of the finished drink — layered fruit tea pours into a clear cup, your hand squeezes a cream top that spreads across the surface, and sticks a pink label onto the cup; the camera pushes in on the layered texture of cream and tea. 6–8s: first-person handheld toast — you raise the fruit tea from 图片2 up to the lens (as if handing it to the viewer), the cup label clearly visible, background voice 「来一口鲜爽」; the final frame freezes on 图片2. All background voices in a consistent female timbre.

The generated commercial — press play to hear the soundtrack. Official ByteDance tutorial sample.

What is Seedance 2.0?

Seedance 2.0 is ByteDance's video generation family and the workhorse line of the Seedance lineup on Molyin. It ships as three live models: Seedance 2.0, the Standard flagship and the only tier that reaches 4K; Seedance 2.0 Fast, a quicker, cheaper cut of the same look; and Seedance 2.0 Mini, an independent lightweight model — not a speed preset — built for high-volume runs at the lowest cost per clip. All three shoot 4–15 second takes from text, an image or mixed references, with voice, sound effects and lip-synced dialogue generated together with the picture.

The reference system is what makes it a daily driver: a single request can carry up to 9 images, 3 video clips and 3 audio tracks — pinning characters and products, borrowing an action from one clip and a camera move from another, all cited by number straight in the prompt. Spoken dialogue lip-syncs with automatic on-screen subtitles, and the family ships two companion models: Seedance 2.0 Video Edit rewrites an existing clip from a single instruction, and Seedance 2.0 Extend continues footage past its last frame.

Seedance 2.0 is in our AI video generator right now — Standard, Fast and Mini in one dropdown, sharing one credit system with the whole lineup, at some of the lowest per-clip Seedance 2.0 prices anywhere.

Seedance 2.0 vs Seedance 2.0 Fast vs Seedance 2.0 Mini

Three tiers, one family — same modes, same reference system, different price-quality trade-offs. Pick the one that fits the shot.

CapabilitySeedance 2.0Seedance 2.0 FastSeedance 2.0 Mini
Resolution tiers480P / 720P / 1080P / 4K (10-bit H.265)480P / 720P480P / 720P
Clip length4–15 s, continuous slider4–15 s4–15 s
Generation modesText / Image / Reference to videoSame three modesSame three modes
Reference inputsUp to 9 images + 3 clips + 3 audio tracksSame reference systemSame reference system
Credits per second11 / 28 / 65 / 135 (480P–4K)9 / 20 (480P / 720P)6 / 12 (480P / 720P)
Best forFlagship quality — final renders and the only 4K tierQuicker drafts — the Standard look, faster and cheaperVolume runs — the lowest cost per clip in the family

All three tiers are available on Molyin and share one credit balance.

Seedance 2.0 vs models live on Molyin

ModelPositioningRelease DateOn Molyin
Seedance 2.0 familyStandard / Fast / Mini workhorses, up to 4KApr 2026 (API)Live — generate above
Seedance 2.5ByteDance flagship — 30s takes, directed second by secondAug 7, 2026 (API)Live — generate above
FLUX 3Black Forest Labs — storyboard keyframes + take continuationAug 7, 2026 (video API)Live — generate above
Wan 3.0Alibaba's All-in-One model, up to 30s takesAug 6, 2026Live — generate above
MiniMax H3Omni-modal flagship, native 2K + stereoJul 31, 2026 · open weights announcedLive — generate above
HappyHorseExpressive generation + instruction video editingJun 2026 (1.1)Live — generate above
Kling 3.0Element references + native audio, up to 4KFeb 4, 2026Live — generate above
Veo 3.1Google's cinematic model with audioOct 2025Live — generate above

One subscription, one credit balance — every model available above is included.

Seedance 2.0 full specifications

Seedance 2.0 full specifications
ParameterValueNotes
Model familyStandard · Fast · MiniThree live models sharing one input contract. Mini is an independent lightweight model, not a speed preset of Standard.
Generation modesText / Image / Reference to videoThree modes on all three models — switch freely in the generator above.
Resolution tiers480P / 720P / 1080P / 4K4K is exclusive to Seedance 2.0 Standard; Fast and Mini generate 480P and 720P.
Clip length4–15 sAny length on a continuous slider, billed per second — the same range across the family.
Aspect ratios16:9 · 9:16 · 4:3 · 3:4 · 21:9 · 1:1Six ratios across all three models, from widescreen 21:9 to vertical 9:16.
Reference imagesUp to 9Characters, products, style and storyboard references, numbered and cited in the prompt — see the camera turntable sample above.
Reference videosUp to 3Action, camera-move and continuation references — the gallery sample above chains three clips into one take.
Reference audio tracksUp to 3Pin a soundtrack or a voice; pair audio references with an image or video input.
AudioNative voice, SFX and music — toggleableDialogue lip-syncs with automatic on-screen subtitles, as in the office sample above; switch audio off for silent plates.
4K output encoding10-bit H.265Standard-tier 4K exports in 10-bit H.265 — clean gradients at high resolution; some older players may need a transcode.
Credits per second6–135 cr/sFrom Mini at 480P to Standard at 4K — full per-tier pricing in the table below; failed generations are refunded.
Editing & extensionSeedance 2.0 Video Edit & ExtendTwo companion models: rewrite an existing clip from one instruction, or continue it past the last frame — pick them in video edit mode.

Everything listed here is already supported on the site, and stays in sync with the model's latest official capabilities.

Seedance 2.0 pricing on Molyin

Standard, Fast and Mini, billed per second from the same credit balance as every other live model. The numbers below are computed straight from our billing engine — and failed generations are refunded automatically.

AI video model credit rates by resolution — credits per second and per-clip examples
ModelResolutionCredits/sec5s Video5s Video + 3s Reference
480p1155 credits75 credits
720p28140 credits191 credits
1080p65325 credits442 credits
4k135675 credits918 credits
480p945 credits62 credits
720p20100 credits136 credits
480p630 credits41 credits
720p1260 credits82 credits
  • Uploaded reference videos are billed at a 60% rate — a 5s generation with a 10s reference video bills 11 seconds, not 15.

See credit pricing for every model on the full price list →

How to generate with Seedance 2.0

Three steps from idea to a finished take with sound

Pick a Seedance 2.0 model and a mode

Switch to Fast or Mini in the dropdown, and choose text-to-video, image-to-video or reference-to-video.

Write the prompt, pin your references

Describe the subject and its motion, then attach up to 9 images, 3 clips and 3 audio tracks and cite them as @-numbered mentions — exactly like the samples on this page.

Generate and export

Pick a tier from 480P drafts to 4K finals, preview in place, then download the original file.

The core prompt formula

The single pattern behind every good Seedance 2.0 prompt: name the subject and its motion, then layer environment, style, camera language and sound as needed. Straight from our full prompt guide — every sample above is built on it.

[Subject] + [Motion] + optional [Environment] [Aesthetics] [Camera] [Audio] [Constraints]

Required

Subject

The star of the frame — a person, animal, or product. The more specific, the better.

Required

Motion

What the subject does, with body parts and pace spelled out.

Optional

Environment

Setting and mood: place, time of day, lighting, weather.

Optional

Aesthetics

Art direction and image quality: palette, texture, clarity.

Optional

Camera

Framing and camera movement: close-up, push-in, dolly — one move per shot.

Optional

Audio

Voice, sound effects, and background music.

Optional

Constraints

Closing negative instructions: "keep it free of subtitles", "do not generate watermarks", "do not generate logos".

Read the full Seedance 2.0 prompt guide

Seedance 2.0 FAQ

The questions creators actually ask, answered straight.

Start Creating with Molyin Today

Sign up free and turn your first idea into a cinematic video in minutes.