Bytedance Seedance 2.0 Mini Text To Video API Documentation

Bytedance Seedance 2.0 Mini Text To Video API Documentation

Playground

Seedance 2.0 Mini Text to Video is ByteDance’s faster, lower-cost text-to-video model for cinematic multi-shot videos. It generates narrative sequences from text prompts with AI camera control, consistent characters across scenes, 480P / 720P / 1080P / 4K output, 4-15s duration, and flexible aspect ratios. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

ByteDance Seedance 2.0 Mini Text-to-Video generates videos from natural-language prompts, with optional reference images, videos, and audio for stronger visual and motion guidance. Describe the scene, action, camera movement, and mood, then choose aspect ratio, resolution, duration, and audio generation settings.

• Need to generate from a start image instead? Try ByteDance Seedance 2.0 Mini Image-to-Video.

Why Choose This?

• Prompt-based video generation
  Generate videos directly from text prompts describing the scene, action, camera movement, and mood.

• Reference-guided generation
  Add reference images, videos, or audio to guide visual style, characters, scene composition, motion, or audio direction.

• Native audio generation
  Generate synchronized audio together with the output video using generate_audio.

• Flexible aspect ratios
  Supports 16:9, 9:16, 4:3, 3:4, 1:1, and 21:9 for different video layouts.

• Resolution options
  Generate videos in 480p, 720p, 1080p, or 4k.

• Web search option
  Enable enable_web_search when the prompt needs real-time information.

Parameters

How to Use

1. Write your prompt — Describe the scene, subject, action, camera movement, lighting, and mood.
2. Add references optional — Upload reference images, videos, or audio when you want stronger guidance.
3. Choose aspect ratio — Select the output format, such as 16:9, 9:16, or 1:1.
4. Choose resolution — Select 480p, 720p, 1080p, or 4k.
5. Set duration — Choose a duration between 4 and 15 seconds.
6. Configure audio optional — Keep generate_audio enabled for synchronized native audio, or disable it if audio is not needed.
7. Submit — Generate the final video.

Example Prompt

A cinematic night market scene, warm lantern light, people walking through a narrow street, soft camera tracking movement, steam rising from food stalls, realistic atmosphere, gentle ambient motion, rich colors, detailed urban background.

Writing Effective Prompts

Most weak results come from thin prompts. Seedance reads a prompt like a shot list — the more you direct it, the more it delivers. Structure your prompt in three layers.

1. One-sentence summary

Open with a single line that fixes the essentials:

> subject + setting + action + style + camera

A lone astronaut walks across a red desert at dusk, cinematic sci-fi, slow low-angle tracking shot.

2. Shot-by-shot timeline

For anything with a beginning, middle, and end, break the duration into timed segments (1-second granularity). Keep the timeline continuous — no gaps.

> 0-3s: The astronaut crests a dune, silhouetted against the setting sun. Slow push-in. > 3-7s: They stop, turn to face the camera, and raise a hand to shield their eyes. Wind lifts red dust across the frame. > 7-10s: Wide pull-back reveals a half-buried spaceship behind them. Warm rim light, long shadows.

Give each range enough to do but not too much — cramming a dozen actions into three seconds causes cuts or dropped beats.

3. Consistency notes

End with anything that should hold across the whole clip: camera style, lighting, color palette, atmosphere, audio.

Handheld feel throughout, shallow depth of field, warm golden-hour palette. Ambient wind and footsteps only — no music.

Camera language

Write these directly; the model understands them:

• Shot size — extreme wide, wide, medium, medium close-up, close-up
• Movement — push in, pull out, pan, track, follow, orbit, tilt up, handheld shake
• Angle — low angle, overhead, eye-level, first-person
• Techniques — one-shot / long take, dolly zoom, bullet time, speed ramp

For a niche term, add a plain-language gloss: “rack focus: the sharp foreground softens as the background comes into focus.”

Referencing uploaded assets

When you attach images, videos, or audio, bind each one explicitly in the prompt by its upload order — @image1, @video1, @audio1 — and say what it’s for. Don’t rely on labels drawn inside the image itself.

• “The knight in @image1 walks through the castle in @image2.”
• “Refer to @video1 for the camera movement only; keep its shot order.”
• “@image1 and @image2 are Character 1, voiced by @audio1.”

When a reference is already accurate, just point to it — no need to re-describe it in detail.

Negative control

Positive descriptions work best, but you can suppress subtitles and audio:

• “No subtitles.”
• “No BGM — ambient and action sounds only.”
• “No audio.”

Weak vs. strong

Pricing

Without reference videos

Output-only pricing:

Prices scale linearly with output duration from 4 to 15 seconds.

With reference videos

Normalized reference-video duration and output duration use the same per-second rate:

Each reference video is normalized to at least 2 seconds, each clip and the combined input are capped at 15 seconds, and the normalized combined input duration is rounded up to a whole second for billing.

Example: 5 seconds of normalized reference input plus 5 seconds of output costs $0.375 at 480p, $0.75 at 720p, $1.875 at 1080p, or $3.75 at 4k.

Best Use Cases

• Text-to-video generation — Create short videos directly from natural-language prompts.
• Reference-to-video workflows — Use reference images, videos, or audio to guide the generated video.
• Cinematic short clips — Generate videos with camera movement, atmosphere, and synchronized audio.
• Character and style consistency — Use reference images to guide characters, visual style, or scene composition.
• Motion and audio guidance — Use reference videos or audio to guide motion direction or sound style.
• Social media content — Generate vertical, square, landscape, or ultra-wide video formats.
• Creative prototyping — Test scene ideas, visual concepts, and motion directions quickly.

Pro Tips

• Be specific about scene, action, camera movement, lighting, and mood.
• Use reference images when character, style, or composition consistency matters.
• Use reference videos when motion direction is important.
• Use reference audio when the generated video should follow a specific audio style or sound direction.
• Use 480p for quick testing, 720p for standard output, 1080p for higher-resolution results, and 4k for maximum-resolution output.
• Keep generate_audio enabled when you want the output video to include synchronized native audio.
• Choose an aspect ratio based on the final format, such as 9:16 for vertical content or 16:9 for widescreen video.

Notes

• Although this endpoint is named Text-to-Video, it also supports reference-to-video workflows through reference_images, reference_videos, and reference_audios.

Related Models

• ByteDance Seedance 2.0 Mini Image-to-Video — Generate video from a start image and prompt.

Notice: Use @image1, @image2, @audio1, etc. to reference your uploaded assets. The references will stay as plain text—don’t worry.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "16:9",
  "resolution": "720p",
  "duration": 5,
  "enable_web_search": false,
  "generate_audio": true
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/bytedance/seedance-2.0-mini/text-to-video" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters