Use Wan 2.1 via API for Effortless Video Creation

Exploring the Full Power of Wan 2.1

What is Wan 2.1？

Alibaba Cloud has taken a significant step by open-sourcing its Wan2.1 series of AI models for video generation. This initiative reduces accessibility barriers for academics, researchers, and commercial users, enabling them to leverage state-of-the-art AI capabilities without substantial upfront costs. By making these models openly available, Alibaba Cloud promotes innovation and collaboration in the fields of AI and video generation.

Versatile Capabilities of Wan2.1

Wan2.1 is a highly versatile tool, excelling in various tasks related to video and image generation:

• Text-to-Video (T2V): Converts text prompts into videos.
• Image-to-Video (I2V): Generates videos from a single image input.
• Video Editing: Enables AI-powered editing for improved video quality and effects.
• Text-to-Image (T2I): Transforms text descriptions into still images.

A standout feature of Wan2.1 is its ability to generate text in both Chinese and English, making it a pioneer in multilingual video generation. Its robust text generation capabilities enhance its practicality across diverse use cases.

Real-World Accuracy of Wan2.1

The Wan2.1 series stands out for its exceptional accuracy in simulating real-world physics and details. Below are some of the key capabilities that highlight its performance:

a. Hand Details

• The model excels at rendering hands, showcasing accurate details like individual fingers, skin textures, and even accessories like rings.

b. Drawing Animation

• Wan2.1 can transform static drawings into short video clips, making it a powerful tool for animators and digital artists.

c. Physics Simulation

• The model accurately depicts real-world physics. For example, it can render a giraffe hanging upside down from a tree, with the tree branch bending realistically under the weight.

d. Hair Movement

• It accurately renders hair movement in videos, showing individual strands as they move naturally when people turn their heads.

e. Object Interactions

• Wan2.1 can simulate complex object interactions, ensuring that multiple objects interact realistically within the same space.

f. Crowd Rendering

• The model excels at rendering large crowds with each individual remaining distinct, resulting in coherent and detailed scenes.

VBench of Wan 2.1

VBench is designed to deliver valuable insights into the strengths and weaknesses of individual models, facilitating detailed and objective evaluation. These insights not only inform future advancements in video generation but also contribute to improving model performance. To ensure consistency with human perception, VBench integrates human preference annotations, reinforcing its relevance and reliability as a benchmark. The chart below highlights the performance of Wan2.1:

Hardware Requirements of Wan2.1

Wan2.1 offers multiple models tailored for different use cases, each with specific hardware demands. Below is a detailed summary of the hardware requirements for the four Wan2.1 models:

Model Name	Function	Resolution Support	Model Size	Hardware Demand	Recommended GPU
T2V-14B	Text-to-Video (T2V)	480P / 720P	14B	⭐⭐⭐⭐	A100 / RTX 3090 / RTX 4090
I2V-14B-720P	Image-to-Video (I2V)	720P	14B	⭐⭐⭐⭐	A100 / RTX 3090 / RTX 4090
I2V-14B-480P	Image-to-Video (I2V)	480P	14B	⭐⭐⭐	RTX 3090 / RTX 4070 Ti
T2V-1.3B	Text-to-Video (T2V)	Low Resolution	1.3B	⭐⭐	RTX 3060 / RTX 4060 or higher

Run Wan2.1 with an API

Novita AI is an AI cloud platform that offers developers an easy way to deploy AI models using our simple API, while also providing the affordable and reliable GPU cloud for building and scaling.

Novita offers highly competitive pricing in the market.

For example, a Wan 2.1 720P 5-second video costs only $0.4 per video

While a similar video on Replicate costs $1 per video

Step 1: Log In and Access the Model Library

Log in to your account and click on the Model Library button.

Step 2: Choose Your Model

Browse through the available options and select the model that suits your needs.

Step 3: Start Your Free Trial

Begin your free trial to explore the capabilities of the selected model.

Step 4: Get Your API Key

To authenticate with the API, we will provide you with a new API key. Entering the “Settings“ page, you can copy the API key as indicated in the image.

Step 5: Install the API

Install API using the package manager specific to your programming language.

After installation, import the necessary libraries into your development environment. Initialize the API with your API key to start interacting with Novita AI LLM. This is an example of using chat completions API for python users.

import requests

url = "https://api.novita.ai/v3/async/wan-t2v"

payload = {
    "extra": {"webhook": {
            "url": "<string>",
            "test_mode": {
                "enabled": True,
                "return_task_status": "<string>"
            }
        }},
    "model_name": "<string>",
    "width": 123,
    "height": 123,
    "seed": 123,
    "prompt": "<string>",
    "frames": 123
}
headers = {
    "Content-Type": "<content-type>",
    "Authorization": "<authorization>"
}

response = requests.request("POST", url, json=payload, headers=headers)

print(response.text)
  
  

Frequently Asked Questions

Novita AI is the All-in-one cloud platform that empowers your AI ambitions. Integrated APIs, serverless, GPU Instance — the cost-effective tools you need. Eliminate infrastructure, start free, and make your AI vision a reality.

Recommend Reading

• Unleash Your Creativity: YouTube Videos Voiceovers Mastery
• 2024 Youtube Video Notes Taker AI Market and Leading Players
• Transforming Images with Ease: Image to Video AI API