An efficient text-to-video model geared toward fast, cost-effective generation. Ideal for prototyping short narrative clips (5–10 s) with stylistic flexibility and prompt-faithful motion.
import requests
import time
# Step 1: Start video generation
generate_url = "https://api.atlascloud.ai/api/v1/model/generateVideo"
headers = {
"Content-Type": "application/json",
"Authorization": "Bearer $ATLASCLOUD_API_KEY"
}
data = {
"model": "bytedance/seedance-v1-lite-t2v-1080p",
"prompt": "A beautiful sunset over the ocean with gentle waves",
"width": 512,
"height": 512,
"duration": 3,
"fps": 24
}
generate_response = requests.post(generate_url, headers=headers, json=data)
generate_result = generate_response.json()
prediction_id = generate_result["data"]["id"]
# Step 2: Poll for result
poll_url = f"https://api.atlascloud.ai/api/v1/model/prediction/{prediction_id}"
def check_status():
while True:
response = requests.get(poll_url, headers={"Authorization": "Bearer $ATLASCLOUD_API_KEY"})
result = response.json()
if result["data"]["status"] in ["completed", "succeeded"]:
print("Generated video:", result["data"]["outputs"][0])
return result["data"]["outputs"][0]
elif result["data"]["status"] == "failed":
raise Exception(result["data"]["error"] or "Generation failed")
else:
# Still processing, wait 2 seconds
time.sleep(2)
video_url = check_status()Seedance 1.0
A video generation model that creates videos from text prompts and images.
🌟 Core Capabilities
🎬 Video Generation
-
Text-to-Video (T2V): Generate videos from text descriptions
-
Image-to-Video (I2V): Generate videos from static images with optional text prompts
-
Resolution: Outputs 1080p videos
🎥 Motion and Dynamics
-
Wide dynamic range supporting both subtle and large-scale movements
-
Maintains physical realism and stability across motion sequences
-
Handles complex action sequences and multi-agent interactions
🧩 Multi-Shot Support
-
Native multi-shot video generation with narrative coherence
-
Maintains consistency in subjects, visual style, and atmosphere across shot transitions
-
Handles temporal and spatial shifts between scenes
🎨 Style and Aesthetics
-
Supports diverse visual styles: photorealism, cyberpunk, illustration, felt texture, and more
-
Interprets stylistic prompts accurately
-
Maintains cinematic quality with rich visual details
🧠 Prompt Understanding
-
Parses natural language descriptions effectively
-
Stable control over camera movements and positioning
-
Accurate interpretation of complex scene descriptions
-
Strong prompt adherence across generated content
⚙️ Technical Specifications
-
Model Version: 1.0
-
Output Resolution: 1080p
-
Input Types: Text prompts, images (for I2V mode)
-
Video Length: Multi-shot capable (specific duration limits not specified)
📊 Model Performance
Based on internal benchmarks using SeedVideoBench-1.0 and third-party evaluations:
-
High scores in prompt adherence
-
Strong motion quality ratings
-
Competitive aesthetic quality
-
Effective source image consistency in I2V tasks
🎯 Use Cases
-
Creative video content generation
-
Prototype development for film and animation
-
Commercial video production
-
Educational and documentary content
-
Fantasy and surreal video creation
⚠️ Limitations
-
Performance metrics based on specific benchmark datasets
-
Actual generation quality may vary with prompt complexity
-
Multi-shot consistency depends on prompt clarity and scene descriptions
