Bytedance Seedance V1.5 Pro Text To Video

Bytedance Seedance V1.5 Pro Text To Video

Playground

Seedance 1.5 Pro (Text-to-Video) generates cinematic, live-action–leaning clips from text with strong prompt adherence, expressive motion, and stable aesthetics. It supports 4–12s duration control (including Smart Duration), multiple aspect ratios (including adaptive), and reproducible generation via seeds—ideal for ads and short-drama workflows.

Features

bytedance/seedance-1.5-pro-t2v

Seedance 1.5 Pro (T2V) is ByteDance Seed’s production-oriented text-to-video model built for cinematic realism, strong prompt adherence, and high expressive motion. It is designed for ad creatives and short-drama workflows where aesthetic stability, emotion-rich acting, and controllable duration matter.

Key Features

• High-quality prompt alignment
  Accurately follows detailed scene descriptions, shot instructions, and emotional tone, producing coherent clips with stable subject and scene intent.

• Expressive motion and emotion
  Natural motion amplitude and strong rhythm; fine-grained facial expressions and emotional delivery, suitable for drama-like performances and high-impact ads.

• Strong aesthetics with live-action realism
  A “natural, live-action” leaning look with stable lighting, composition, and color harmony for premium-looking outputs.

• Rich effects via prompting
  The base model supports a wide variety of visual effects achievable directly through prompting, enabling template-like creative effects without extra tooling.

• Flexible duration control (4–12 seconds)
  Specify any integer duration from 4s to 12s. Supports Smart Duration (-1) where the model selects an appropriate length within the same range.

• Aspect ratio control
  Choose either adaptive (model selects the best ratio from the prompt) or explicitly set one of:
  16:9, 9:16, 4:3, 3:4, 1:1, 21:9.

Parameters

• prompt (required)
  Text description of the scene, style, subject actions, camera motion, and overall mood.

• duration
  Integer seconds in [4, 12]. Use -1 for Smart Duration (model decides within [4, 12]).

• resolution
  One of 480p, 720p.

• fps
  Fixed at 24.

• ratio
  Use adaptive (model selects a suitable ratio) or set one of: 16:9, 9:16, 4:3, 3:4, 1:1, 21:9.

• fixed_camera
  Boolean. If true, the camera remains fixed; if false, camera motion is driven by the prompt.

• seed
  Integer. Controls randomness; the same seed tends to produce more similar outputs. Set -1 to cancel random seed.

Use Cases

• Performance-heavy ad creatives: emotion-forward acting beats with stable visual quality.
• Live-action style short drama clips: realistic lighting/composition and coherent storytelling moments.
• Template-like VFX content: prompt-driven effects for fast experimentation and iteration.

Notes

• For the cleanest motion, keep prompts specific but not overloaded: limit the number of primary subjects and simultaneous actions, especially in short clips.
• If you see shaky composition or drifting framing, set camera_fixed = true and describe motion using subject movement rather than camera movement.
• When using duration = -1, write prompts with a clear narrative “end beat” (a natural stopping point) so the model has a reason to conclude.

Related Models

• ByteDance Seedance V1 Pro Fast (Text-to-Video) – Faster Seedance variant for quick iterations and previews.
• ByteDance Seedance V1 Pro (T2V 480p) – A lower-res option for budget-friendly testing and rapid prompt iteration.
• WaveSpeedAI HunyuanVideo-1.5 (Text-to-Video) – Lightweight T2V model with strong motion coherence and efficient inference.
• WaveSpeedAI HunyuanVideo (T2V) – Another T2V option for broader experimentation across styles and pacing.

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/bytedance/seedance-v1.5-pro/text-to-video" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "aspect_ratio": "16:9",
    "duration": 5,
    "resolution": "720p",
    "generate_audio": true,
    "camera_fixed": false,
    "seed": -1
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"