Seedance 1.0 lite i2v
Seedance 1.0 transforms text and images into engaging 720p dynamic videos with cinematic storytelling.
API
If you're looking for an API, you can choose from your desired programming language.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
import requests
import base64
# Use this function to convert an image file from the filesystem to base64
def image_file_to_base64(image_path):
with open(image_path, 'rb') as f:
image_data = f.read()
return base64.b64encode(image_data).decode('utf-8')
# Use this function to fetch an image from a URL and convert it to base64
def image_url_to_base64(image_url):
response = requests.get(image_url)
image_data = response.content
return base64.b64encode(image_data).decode('utf-8')
# Use this function to convert a list of image URLs to base64
def image_urls_to_base64(image_urls):
return [image_url_to_base64(url) for url in image_urls]
api_key = "YOUR_API_KEY"
url = "https://api.segmind.com/v1/seedance-v1-lite-image-to-video"
# Request payload
data = {
"image_url": "https://segmind-resources.s3.amazonaws.com/input/73a8dadb-95e6-4bdc-a3ae-a4cd041b99a0-b572c379-3eb8-4c67-893c-b08662f9d11f.jpeg",
"duration": 5,
"prompt": "Generate a serene forest scene at dawn with birds chirping, sunlight filtering through trees, and a gentle mist.",
"resolution": "720p",
"seed": 12345,
"camera_fixed": False
}
headers = {'x-api-key': api_key}
response = requests.post(url, json=data, headers=headers)
print(response.content) # The response is the generated imageAttributes
URL to the first frame image for video generation (legacy parameter). The model will use this image as the starting frame to generate the video based on your prompt. Note: Cannot be used together with first_frame_url or last_frame_url. For new integrations, prefer using first_frame_url instead.
URL to the first frame image for video generation. Use this to specify the starting frame of your video. Can be used alone or together with last_frame_url. Note: Cannot be used together with image_url (reference image).
URL to the last frame image for video generation. Use this together with first_frame_url to specify both the starting and ending frames of your video. The model will generate smooth transitions between the two frames. Note: Requires first_frame_url to be provided.
Set video length in seconds. Opt for 5 seconds for quick previews, 10 for detailed scenes.
Allowed values:
Detail the animation scene vividly. E.g., A sunset beach scene with waves lapping, people playing, sky turning orange.
Choose video clarity; 720p for most uses, 480p for faster processing.
Allowed values:
Define a seed for consistent outputs. Any number 1-999999 suffices.
min : 1,
max : 999999
Whether to fix the camera position.
To keep track of your credit usage, you can inspect the response headers of each API call. The x-remaining-credits property will indicate the number of remaining credits in your account. Ensure you monitor this value to avoid any disruptions in your API usage.
Resources to get you started
Everything you need to know to get the most out of Seedance 1.0 lite i2v
# Seedance 1.0 Usage Guide
Seedance 1.0 is a text-to-video engine that transforms your prompts and reference images into coherent multi-shot videos at up to 720p. Follow this guide to tune parameters, optimize quality, and match your workflow—whether you need a quick preview or a polished cinematic cutscene.
## 1. Core Inputs
- **prompt** (required): A vivid description of scene, action, style and camera moves.
- **image_url** (required): S3 link to your reference or relighting image.
- **duration** (required): 5s or 10s.
- **resolution** (advanced): 480p (fast) or 720p (detailed).
- **seed** (advanced, optional): Integer (1–999,999) to reproduce results.
## 2. Prompting Best Practices
1. **Be Specific**
- Setting: “Sunset beach with rolling waves.”
- Action: “Surfers riding close to camera.”
- Camera: “Wide-to-close pan over the shoreline.”
2. **Style & Lighting**
- “Photorealistic dawn mist,” “Cyberpunk neon reflections on wet asphalt.”
3. **Multi-Shot Sequencing**
- Use “dolly in,” “track left,” “cut to close-up.”
4. **Duration Choice**
- 5s: Fast A/B tests, thumbnails, previews.
- 10s: Narrative beats, longer camera moves, stable storytelling.
5. **Consistency Across Runs**
- Fix the same seed for multiple variants or shots.
## 3. Parameter Presets by Use Case
| Use Case | Duration | Resolution | Seed | Prompt Style |
|-----------------------------|----------|------------|---------|-------------------------------------|
| Quick Preview | 5s | 480p | (none) | “Midday forest clearing, static cam”|
| Social Media Teaser | 5s | 720p | 42 | “Close-up product spin, soft lighting” |
| Storyboard & Prototyping | 10s | 480p | 2023 | “Two characters talk, over-shoulder cut” |
| Game Cinematic Sequence | 10s | 720p | 999999 | “Epic battle, wide shot → tight focus” |
| Mood-Board Concept Art | 10s | 720p | 12345 | “Neo-Tokyo alley, rain, neon signs” |
## 4. Tips to Maximize Quality
- **Shorter Resolutions** reduce rendering time but lower detail.
- **Higher Seeds** produce subtle variations; lower seeds are bolder.
- **Iterate Quickly**: Lock prompt structure, swap only one variable per test.
- **Combine Styles**: Photorealism + stylized color grading for unique flair.
---
By following these guidelines and parameter presets, you’ll achieve stable frames, accurate prompt fidelity, and cinematic polish with Seedance 1.0—whether creating a rapid storyboard or a full 10-second hero sequence.
Other Popular Models
Discover other models you might be interested in.
illusion-diffusion-hq
Monster Labs QrCode ControlNet on top of SD Realistic Vision v5.1
sd1.5-majicmix
The most versatile photorealistic model that blends various models to achieve the amazing realistic images.
codeformer
CodeFormer is a robust face restoration algorithm for old photos or AI-generated faces.
sd2.1-faceswapper
Take a picture/gif and replace the face in it with a face of your choice. You only need one image of the desired face. No dataset, no training
