Przejdź do treści

Seedance 2.0 Fast Image to Video

Integracja

Identyfikator modelu bytedance/seedance-2.0/fast/image-to-video
# Run the model; wait up to 20 s for the finished result in the same call
curl -X POST "https://api.doitong.com/v1/run/bytedance/seedance-2.0/fast/image-to-video?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Create an exactly 15-second, 16:9 Full HD (1920×1080), 24 FPS photorealistic cinematic beauty film using the uploaded reference image and character sheet. Preserve the character’s exact identity, facial features, hairstyle, hair color, outfit, body proportions, environment, and original visual aesthetic.\n\nBegin with a high-angle shot of the character lying diagonally on the floor among scattered sheet music, gazing blankly at the notes. Slowly glide the camera toward her face as a loose page drifts across the lens, creating a paper-wipe transition. Cut to an extreme close-up of her eyes, eyelashes, and natural skin texture, with one subtle blink, a distant gaze, and a slow macro push-in. Match-cut from the reflection in her eye to a sheer curtain gently billowing in cool daylight, with subtle movement in nearby music sheets and loose hair strands.\n\nTransition through the moving curtain into a low-angle, floor-level tracking shot, moving from her hand holding sheet music along her resting forearm toward her face. She must remain completely flat on the floor, with her cheek resting on her forearm and that arm lying directly on the floor. Her free hand loosely holds a sheet of music near her face. Only her eyes shift toward the camera; her head, shoulders, and upper body remain resting on the floor. Maintain her melancholic, distant expression.\n\nFor the final four seconds, reveal a wider cinematic composition of the character, scattered sheet music, softly moving curtain, and existing environmental elements. Slowly pull back while drifting sideways, allowing a few pages to move naturally in the breeze. Hold the final beauty frame briefly and fade smoothly to black at exactly 15 seconds.\n\nUse elegant editorial cinematography, soft diffused lighting, natural skin texture, delicate facial highlights, expressive eyes, shallow depth of field, subtle film grain, muted pastel tones, refined cinematic color grading, and graceful camera movement. Preserve the reference image’s original lighting and color palette, including warm amber highlights and cool blue daylight if present. Create five distinct shots with intentional paper-wipe, eye-reflection, curtain-wipe, and gaze-match-cut transitions. Keep character animation minimal and environmental motion realistic.\n\nNegative prompt: Identity drift, altered face, hairstyle, hair color, outfit, or environment; extra characters; smiling; exaggerated expressions; raised head; lifted shoulders; elbow-supported pose; elevated upper body; sitting; unnatural neck angles; deformed hands; extra fingers; unnatural paper grip; floating pages; chaotic wind; excessive hair movement; unrealistic physics; jitter; flicker; shaky camera; random zooms; abrupt transitions; harsh lighting; plastic skin; excessive retouching; cartoon; anime; vertical framing; text; logos; watermarks; low resolution.\n\nCritical requirement: Create a true five-shot cinematic beauty film, not a single continuous shot with simple zooms. Preserve the character’s identity and original design. Throughout the film, she must remain lying flat, with her cheek resting on an arm directly on the floor and her free hand holding sheet music. Never raise her head or upper body. Prioritize expressive eyes, elegant camera movement, creative transitions, and melancholic beauty. Exactly 15 seconds, 16:9 horizontal, Full HD 1920×1080.",
    "image": "https://content.doitong.com/dev-api/2026/10/21/req_6qbgnre4o2i6n5gbq08x9fr9/0-yJGsQzt9-eb365egOeXAS_NiQledyV.png",
    "resolution": "480p",
    "duration": "5",
    "aspect_ratio": "16:9",
    "generate_audio": true
  }'

# Still "queued" or "running"? Poll with the id from the response:
curl "https://api.doitong.com/v1/requests/REQUEST_ID?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY"

To żądanie: 0,565 $ 565 kredytów API Uruchom na tej stronie

Uzyskaj klucz API
ByteDance Obraz na wideo Natywne audioSzybki4–15 sdo 720p #fast#first-last-frame

Szybszy i tańszy Seedance 2.0 image-to-video w 480p lub 720p z dźwiękiem.

Przykłady

Dane wejściowe
Wynik · ~173 s

PromptCreate an exactly 15-second, 16:9 Full HD (1920×1080), 24 FPS photorealistic cinematic beauty film using the uploaded reference image and character sheet. Preserve the character’s exact identity, facial features, hairstyle, hair color, outfit, body proportions, environment, and original visual aesthetic. Begin with a high-angle shot of the character lying diagonally on the floor among scattered sheet music, gazing blankly at the notes. Slowly glide the camera toward her face as a loose page drifts across the lens, creating a paper-wipe transition. Cut to an extreme close-up of her eyes, eyelashes, and natural skin texture, with one subtle blink, a distant gaze, and a slow macro push-in. Match-cut from the reflection in her eye to a sheer curtain gently billowing in cool daylight, with subtle movement in nearby music sheets and loose hair strands. Transition through the moving curtain into a low-angle, floor-level tracking shot, moving from her hand holding sheet music along her resting forearm toward her face. She must remain completely flat on the floor, with her cheek resting on her forearm and that arm lying directly on the floor. Her free hand loosely holds a sheet of music near her face. Only her eyes shift toward the camera; her head, shoulders, and upper body remain resting on the floor. Maintain her melancholic, distant expression. For the final four seconds, reveal a wider cinematic composition of the character, scattered sheet music, softly moving curtain, and existing environmental elements. Slowly pull back while drifting sideways, allowing a few pages to move naturally in the breeze. Hold the final beauty frame briefly and fade smoothly to black at exactly 15 seconds. Use elegant editorial cinematography, soft diffused lighting, natural skin texture, delicate facial highlights, expressive eyes, shallow depth of field, subtle film grain, muted pastel tones, refined cinematic color grading, and graceful camera movement. Preserve the reference image’s original lighting and color palette, including warm amber highlights and cool blue daylight if present. Create five distinct shots with intentional paper-wipe, eye-reflection, curtain-wipe, and gaze-match-cut transitions. Keep character animation minimal and environmental motion realistic. Negative prompt: Identity drift, altered face, hairstyle, hair color, outfit, or environment; extra characters; smiling; exaggerated expressions; raised head; lifted shoulders; elbow-supported pose; elevated upper body; sitting; unnatural neck angles; deformed hands; extra fingers; unnatural paper grip; floating pages; chaotic wind; excessive hair movement; unrealistic physics; jitter; flicker; shaky camera; random zooms; abrupt transitions; harsh lighting; plastic skin; excessive retouching; cartoon; anime; vertical framing; text; logos; watermarks; low resolution. Critical requirement: Create a true five-shot cinematic beauty film, not a single continuous shot with simple zooms. Preserve the character’s identity and original design. Throughout the film, she must remain lying flat, with her cheek resting on an arm directly on the floor and her free hand holding sheet music. Never raise her head or upper body. Prioritize expressive eyes, elegant camera movement, creative transitions, and melancholic beauty. Exactly 15 seconds, 16:9 horizontal, Full HD 1920×1080.

5 s480p16:9generate_audio: true0,565 $ · 565 kredytów API

Prompt od @ariaxawan na X

Zbuduj żądanie

Elementy sterujące są generowane ze schematu modelu. Kod i cena aktualizują się podczas ich zmiany.

•wymagane

Instrukcje dotyczące ruchu i działania wideo podane jako prompt tekstowy.

•wymagane image_url
Upuść plik lub

URL początkowego obrazu, który zostanie animowany. Akceptowane formaty to JPEG, PNG i WebP; limit rozmiaru wynosi 30 MB.

duration 5 s

Wybór długości wideo: `auto` przekazuje wybór modelowi na podstawie promptu, natomiast jawny czas trwania musi wynosić od 4 do 15 sekund.

resolution

Rozdzielczość wyniku: 480p zapewnia szybkość generowania, a 720p stanowi zrównoważony wybór.

aspect_ratio

Proporcje obrazu wyniku: 16:9 oznacza orientację poziomą, 9:16 pionową, 1:1 kwadrat, a 21:9 kinowy format ultrapanoramiczny. Wybierz `auto`, aby wyprowadzić proporcje z obrazu wejściowego.

Zaawansowane end_image · bitrate_mode · codec
end_image_url
Upuść plik lub

URL wskazujący obraz JPEG, PNG lub WebP o rozmiarze nie większym niż 30 MB, definiujący ostatnią klatkę wideo. Podanie tego obrazu sprawia, że staje się on miejscem docelowym przejścia generowanego wideo rozpoczynającego się od obrazu początkowego.

bitrate_mode domyślnie standard

Wybierz szybkość transmisji kodowania wyniku: 'standard' zachowuje domyślną wartość, natomiast 'high' nakazuje modelowi poprawić jakość kodowania kosztem większego pliku.

codec domyślnie auto

Wybór kodeka wideo: `auto` pozostawia domyślne działanie kodeka bez zmian, `H264` wybiera H.264, a `H265` wybiera H.265.

POST /run/bytedance/seedance-2.0/fast/image-to-video?wait=20 wycena → 0,565 $
# Run the model; wait up to 20 s for the finished result in the same call
curl -X POST "https://api.doitong.com/v1/run/bytedance/seedance-2.0/fast/image-to-video?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "Create an exactly 15-second, 16:9 Full HD (1920×1080), 24 FPS photorealistic cinematic beauty film using the uploaded reference image and character sheet. Preserve the character’s exact identity, facial features, hairstyle, hair color, outfit, body proportions, environment, and original visual aesthetic.\n\nBegin with a high-angle shot of the character lying diagonally on the floor among scattered sheet music, gazing blankly at the notes. Slowly glide the camera toward her face as a loose page drifts across the lens, creating a paper-wipe transition. Cut to an extreme close-up of her eyes, eyelashes, and natural skin texture, with one subtle blink, a distant gaze, and a slow macro push-in. Match-cut from the reflection in her eye to a sheer curtain gently billowing in cool daylight, with subtle movement in nearby music sheets and loose hair strands.\n\nTransition through the moving curtain into a low-angle, floor-level tracking shot, moving from her hand holding sheet music along her resting forearm toward her face. She must remain completely flat on the floor, with her cheek resting on her forearm and that arm lying directly on the floor. Her free hand loosely holds a sheet of music near her face. Only her eyes shift toward the camera; her head, shoulders, and upper body remain resting on the floor. Maintain her melancholic, distant expression.\n\nFor the final four seconds, reveal a wider cinematic composition of the character, scattered sheet music, softly moving curtain, and existing environmental elements. Slowly pull back while drifting sideways, allowing a few pages to move naturally in the breeze. Hold the final beauty frame briefly and fade smoothly to black at exactly 15 seconds.\n\nUse elegant editorial cinematography, soft diffused lighting, natural skin texture, delicate facial highlights, expressive eyes, shallow depth of field, subtle film grain, muted pastel tones, refined cinematic color grading, and graceful camera movement. Preserve the reference image’s original lighting and color palette, including warm amber highlights and cool blue daylight if present. Create five distinct shots with intentional paper-wipe, eye-reflection, curtain-wipe, and gaze-match-cut transitions. Keep character animation minimal and environmental motion realistic.\n\nNegative prompt: Identity drift, altered face, hairstyle, hair color, outfit, or environment; extra characters; smiling; exaggerated expressions; raised head; lifted shoulders; elbow-supported pose; elevated upper body; sitting; unnatural neck angles; deformed hands; extra fingers; unnatural paper grip; floating pages; chaotic wind; excessive hair movement; unrealistic physics; jitter; flicker; shaky camera; random zooms; abrupt transitions; harsh lighting; plastic skin; excessive retouching; cartoon; anime; vertical framing; text; logos; watermarks; low resolution.\n\nCritical requirement: Create a true five-shot cinematic beauty film, not a single continuous shot with simple zooms. Preserve the character’s identity and original design. Throughout the film, she must remain lying flat, with her cheek resting on an arm directly on the floor and her free hand holding sheet music. Never raise her head or upper body. Prioritize expressive eyes, elegant camera movement, creative transitions, and melancholic beauty. Exactly 15 seconds, 16:9 horizontal, Full HD 1920×1080.",
    "image": "https://content.doitong.com/dev-api/2026/10/21/req_6qbgnre4o2i6n5gbq08x9fr9/0-yJGsQzt9-eb365egOeXAS_NiQledyV.png",
    "resolution": "480p",
    "duration": "5",
    "aspect_ratio": "16:9",
    "generate_audio": true
  }'

# Still "queued" or "running"? Poll with the id from the response:
curl "https://api.doitong.com/v1/requests/REQUEST_ID?wait=20" \
  -H "Authorization: Bearer $DOITONG_API_KEY"

Uruchom to żądanie

Klucz pozostaje w tej przeglądarce: jest wysyłany tylko do Doitong API i zapisywany (localStorage) tylko po zaznaczeniu opcji Zapamiętaj. Uzyskaj klucz API

Utworzone w Doitong

Publiczne prace utworzone za pomocą Seedance 2.0 przez społeczność Doitong. Użyj ponownie promptu jednym kliknięciem.

Pokazane publiczne prace: 16
Subject: Man in a bunny suit, frozen city crowd Context: Urban street, all people, vehicles, and ... Wideo

Subject: Man in a bunny suit, frozen city crowd Context: Urban street, all people, vehicles, and trees are frozen mid-action Action: Only the bunny-suited man moves, walking and waving in front of an unresponsive man in a black suit Style: Cinematic, intense stillness, Quicksilver/X-Men, Matrix...

This is a cinematic scene filmed with a hand-held shaky camera. The entire world is completely fr... Wideo

This is a cinematic scene filmed with a hand-held shaky camera. The entire world is completely frozen in time. Nothing moves. Absolutely nothing. All people are frozen, cars are frozen, bicycles are frozen, traffic lights are frozen, trees and leaves are frozen, wind is frozen, shadows are...

A close-up portrait of a striking cyberpunk woman with sleek dark hair partially wet and falling ... Wideo

A close-up portrait of a striking cyberpunk woman with sleek dark hair partially wet and falling around her face. She’s wearing glossy, high-tech neon visor goggles with a vivid color gradient — bright yellow, orange, and pink reflecting glowing city lights. Her expression is confident and...

Ultra-realistic scene featuring miniature construction workers in orange vests and hard hats, act... Wideo

Ultra-realistic scene featuring miniature construction workers in orange vests and hard hats, actively working on a giant [pizza]. The tiny workers are using small tools, cranes, and machines to [assemble]. The [pizza] is placed on a metal surface or kitchen countertop. Steam or particles rise...

Аня вздрагивает от трезвона десяти будильников, резко садится в кровати, хлопает по ним руками в ... Wideo

Аня вздрагивает от трезвона десяти будильников, резко садится в кровати, хлопает по ним руками в попытке выключить, но они продолжают пищать, она в панике оглядывается. Камера медленно приближается с низкого угла от ног кровати к лицу, создавая ощущение нарастания хаоса. Лучи утреннего солнца...

Анна мягко гладит спящую собаку и смотрит на заходящее солнце с усталой, но счастливой улыбкой. К... Wideo

Анна мягко гладит спящую собаку и смотрит на заходящее солнце с усталой, но счастливой улыбкой. Камера медленно отъезжает назад, охватывая весь уютный дом и сад. Свет солнца постепенно затухает, переходя в сумерки. Финальный кадр передает чувство семейного тепла и завершенности прекрасного дня....

Аня падает на корточки у двери, лихорадочно надевает разные носки, роняет помаду из сумки, вскаки... Wideo

Аня падает на корточки у двери, лихорадочно надевает разные носки, роняет помаду из сумки, вскакивает, хватает ключи. Камера кружит вокруг нее на 180 градусов на уровне колен, ускоряясь. Ключи звенят, сумка болтается. Эмоциональный бит — пик хаоса с ноткой абсурдного юмора. Dialogue in this...

Буня несется во весь опор, виляя хвостом, сосиска забавно болтается у него в пасти. Камера быстро... Wideo

Буня несется во весь опор, виляя хвостом, сосиска забавно болтается у него в пасти. Камера быстро следует за ним на уровне земли, создавая эффект скорости. Анна и дети вбегают в кадр и выбегают из него, размахивая руками. Пыль слегка поднимается из-под лап. Сцена наполнена комичным драйвом и...

Informacje o Seedance 2.0 Fast Image to Video

Szybki wariant Seedance 2.0 animuje obraz początkowy, opcjonalnie przechodząc do obrazu końcowego, z zsynchronizowanym audio, w 480p lub 720p. 4–15 sekund albo długość automatyczna. Służy do taniego testowania pomysłów na ruch przed renderowaniem w pełnej jakości.

Parametry

Wyślij nazwę kanoniczną albo natywną, nie obie. Nazwy kanoniczne

NazwaTyp DomyślneDozwolone Opis
promptwymagane
Typstring
Instrukcje dotyczące ruchu i działania wideo podane jako prompt tekstowy.
image_urlwymagane image
Typurl · image
URL początkowego obrazu, który zostanie animowany. Akceptowane formaty to JPEG, PNG i WebP; limit rozmiaru wynosi 30 MB.
end_image_urlopcjonalne end_image
Typurl · image
URL wskazujący obraz JPEG, PNG lub WebP o rozmiarze nie większym niż 30 MB, definiujący ostatnią klatkę wideo. Podanie tego obrazu sprawia, że staje się on miejscem docelowym przejścia generowanego wideo rozpoczynającego się od obrazu początkowego.
resolutionopcjonalne
Typenum
Domyślne720p
480p720p
Rozdzielczość wyniku: 480p zapewnia szybkość generowania, a 720p stanowi zrównoważony wybór.
durationopcjonalne
Typenum
Domyślneauto
auto456789101112131415
Wybór długości wideo: `auto` przekazuje wybór modelowi na podstawie promptu, natomiast jawny czas trwania musi wynosić od 4 do 15 sekund.
aspect_ratioopcjonalne
Typenum
Domyślneauto
auto21:916:94:31:13:49:16
Proporcje obrazu wyniku: 16:9 oznacza orientację poziomą, 9:16 pionową, 1:1 kwadrat, a 21:9 kinowy format ultrapanoramiczny. Wybierz `auto`, aby wyprowadzić proporcje z obrazu wejściowego.
generate_audioopcjonalne
Typboolean
Domyślnetrue
Włącz, aby dołączyć do wideo zsynchronizowane audio. Generowane audio zawiera mowę dopasowaną do ruchu ust, atmosferę otoczenia i efekty dźwiękowe.
bitrate_modeopcjonalne
Typenum
Domyślnestandard
standardhigh
Wybierz szybkość transmisji kodowania wyniku: 'standard' zachowuje domyślną wartość, natomiast 'high' nakazuje modelowi poprawić jakość kodowania kosztem większego pliku.
codecopcjonalne
Typenum
Domyślneauto
autoH264H265
Wybór kodeka wideo: `auto` pozostawia domyślne działanie kodeka bez zmian, `H264` wybiera H.264, a `H265` wybiera H.265.
wymagane Nazwa kanoniczna, również akceptowana

Parametry plików przyjmują publiczny URL https://, identyfikator URI data: do 10 MB albo file_url przesłanego pliku. Pliki

Wynik

GET /requests/req_8k2m4q7x1z9c3v5b6n0p2r4t
{
  "id": "req_8k2m4q7x1z9c3v5b6n0p2r4t",
  "object": "request",
  "model": "bytedance/seedance-2.0/fast/image-to-video",
  "status": "succeeded",
  "created_at": "2026-10-20T12:00:00Z",
  "started_at": "2026-10-20T12:00:02Z",
  "completed_at": "2026-10-20T12:00:09Z",
  "queue_position": null,
  "cost": {
    "credits": 565,
    "usd": "0.565",
    "charged": 565,
    "refunded": 0
  },
  "input": {
    "prompt": "The fox stops, looks up at the fireflies and leaps to catch one, playful mood",
    "image": "https://content.doitong.com/images/showcase/v3/gen_fox_cartoon.jpg",
    "duration": 5,
    "resolution": "480p"
  },
  "output": {
    "video": {
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.mp4",
      "content_type": "video/mp4",
      "file_name": "0-output.mp4",
      "file_size": 1843200
    },
    "seed": 42
  },
  "outputs": [
    {
      "type": "video",
      "url": "https://content.doitong.com/dev-api/2026/10/20/req_8k2m4q7x1z9c3v5b6n0p2r4t/0-output.mp4",
      "content_type": "video/mp4",
      "file_size": 1843200,
      "width": 1280,
      "height": 720,
      "duration": 5,
      "stored": true
    }
  ],
  "error": null,
  "metrics": {
    "inference_seconds": 6.8
  },
  "urls": {
    "self": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t",
    "cancel": "https://api.doitong.com/v1/requests/req_8k2m4q7x1z9c3v5b6n0p2r4t/cancel"
  }
}
Twoje pliki, Twój wybórKażdy outputs[].url to kopia w naszej przestrzeni dyskowej: przechowywana bezterminowo (opłata za GB-miesiąc) lub 7 dni bezpłatnie, zgodnie z ustawieniem konta lub ?storage=.
Taki sam format w każdym modelutype, url, content_type, width, height i duration — tak samo dla obrazów, filmów i dźwięku. output to własny obiekt wyniku modelu.
Pola wyniku
video
object {url, content_type, file_name, file_size}Plik wideo utworzony podczas generowania.
seed
integerWartość ziarna stosowana podczas generowania.

Więcej modeli: Obraz na wideo

Dla agentów AI: ta strona w formacie Markdown jest dostępna pod adresem https://doitong.com/pl/developers/models/bytedance/seedance-2.0/fast/image-to-video.md. Wszystkie modele i przewodniki: https://doitong.com/developers/llms.txt