Seedance 1.5 Pro Text to Video Fast | Cinematic AI Text to Video (4–12s)

ByteDance Seedance V1.5 Pro Text-to-Video Fast

Seedance V1.5 Pro Text-to-Video Fast is ByteDance’s production-oriented text-to-video model on WaveSpeedAI. It generates short, cinematic clips from a single natural-language prompt with strong prompt adherence and expressive motion, optimized for fast iteration across common aspect ratios.

Key capabilities

Text-to-video generation from one prompt for fast concept-to-clip workflows
Strong prompt adherence for subject, action, environment, lighting, and mood
Expressive motion with stable aesthetics suited to cinematic short-form outputs
Aspect ratio options for vertical, square, and widescreen delivery
Optional audio generation with a simple toggle
Seed control for reproducible iterations and controlled variations

Parameters

Parameter	Description
prompt*	The positive prompt describing what to generate (subject + action + scene + mood).
aspect_ratio	Output aspect ratio: 21:9, 16:9, 4:3, 1:1, 3:4, 9:16.
duration	Output duration in seconds (available values depend on the endpoint UI/validator).
resolution	Output resolution: 720p or 1080p.
generate_audio	Whether to generate audio (true/false).
camera_fixed	Whether to lock camera position (true/false).
seed	Random seed (-1 for random; set a fixed integer for repeatable results).

How to use

Write a clear prompt describing the subject, action, and scene.
Choose an aspect_ratio that matches your target platform (9:16 for shorts, 16:9 for YouTube, 1:1 for feeds).
Set duration (start short for iteration, then increase once you like the look).
Pick resolution:
- 720p for faster drafts
- 1080p for final output detail
Toggle generate_audio on if you want sound included, or off if you’ll add audio in post.
Use camera_fixed if you want a locked shot and less framing drift.
Set seed for reproducible reruns, or keep -1 for fresh variations.

Prompt tips

Write prompts like a director’s brief:

Subject: who/what is on screen
Action: what happens next (use clear verbs)
Scene: where it happens + time + lighting
Camera: framing and movement (only if needed)
Style: mood and visual language (optional)

Practical tips:

Prefer simple, direct sentences first. Add detail only when you need tighter control.
This workflow works best with positive instructions (describe what you want, not long avoid-lists).
If you want stronger motion, use intensity words like subtle, energetic, rapid, slow, high energy.
For multi-step actions, list them in order: walks in → pauses → looks up → smiles.

Pricing

Effective rates

Resolution	Generate audio	Price per 5 seconds	Price per second
720p	true	$0.20	$0.04
720p	false	$0.10	$0.02
1080p	true	$0.30	$0.06
1080p	false	$0.15	$0.03

Example costs

Duration	Resolution	Generate audio	Cost per run
5s	720p	true	$0.20
5s	720p	false	$0.10
12s	720p	true	$0.48
5s	1080p	true	$0.30
5s	1080p	false	$0.15
12s	1080p	true	$0.72

Notes

Start with 720p and shorter duration for quick iteration, then switch to 1080p for finals.
If framing drifts, try camera_fixed = true and move the action onto the subject instead of the camera.
Keeping the same seed makes small prompt changes easier to compare.

Related Models

Seedance V1.5 Pro (Text-to-Video) — Non-fast variant for steadier outputs when speed is less critical.
Seedance V1 Pro Fast (Text-to-Video) — Fast alternative for rapid previews and lightweight iteration.
HunyuanVideo-1.5 (Text-to-Video) — Efficient T2V model with strong motion coherence for cost-effective generation.
Alibaba Wan 2.5 (Text-to-Video Fast) — Fast T2V option for broad style exploration and high-volume testing.

示例查看全部

README