Wan I2V 480p Ultra Fast

The Wan2.1 14B model is an advanced image-to-video model that offers accelerated inference capabilities, enabling high-res video generation with high visual quality and motion diversity

Features

Wan2.1-i2v-480p-ultra-fast is an open-source AI video generation model developed by Alibaba Cloud, designed for image-to-video tasks. The 14-billion-parameter professional version excels in generating complex motions and simulating physical dynamics, delivering exceptional performance.

Built upon a causal 3D Variational Autoencoder (VAE) and Video Diffusion Transformer architecture, Wan2.1-i2v-480p-ultra-fast efficiently models spatiotemporal dependencies. In the authoritative VBench evaluation, the 14B version achieved a leading score of 86.22%, surpassing models like Sora, Luma, and Pika, and securing the top position.

The model is available on Wavespeed AI, providing convenient access for developers.Leveraging cutting-edge acceleration techniques, wan-2.1/i2v-720p-ultra-fast pushes the limits of rapid video synthesis for creative and practical applications.

Key Features

• State-of-the-Art Performance: Maintains top-tier video generation performance by consistently outperforming both open-source and commercial solutions on a variety of benchmarks.
• Ultra-Fast Inference: Enhanced with additional optimization layers, this ultra-fast variant significantly reduces the time required to generate videos. Experience a noticeable reduction in latency compared to the standard model, making real-time applications even more viable.
• Consumer-Grade GPU Compatibility: Optimized for efficient processing on widely available hardware—with support similar to models such as T2V-1.3B which require only 8.19 GB of VRAM. This enables video generation on consumer-grade GPUs without compromising quality.
• Robust Video VAE Integration: Incorporates a powerful video variational autoencoder that efficiently encodes and decodes video content at 1080P resolution, preserving crucial temporal information while generating 480P output.

ComfyUI

wan-2.1/i2v-480p-ultra-fast is also available on ComfyUI, providing local inference capabilities through a node-based workflow, ensuring flexible and efficient image generation on your system.

Limitations

• Creative Rather than Factual: wan-2.1/i2v-480p-ultra-fast is crafted for creative image-to-video generation and is not intended for scenarios that require factually accurate or verifiable content.
• Potential Inherent Biases: Like any data-driven model, it may reflect underlying biases from its training data.
• Prompt Sensitivity: The quality and consistency of the generated video may vary based on the clarity and specificity of the input image.
• Task-Specific Functionality: This model is solely focused on converting images to videos at 480P resolution and does not support additional video generation tasks, such as text-to-video or video editing.

Out-of-Scope Use

The model and its derivatives may not be used in any way that violates applicable national, federal, state, local, or international law or regulation, including but not limited to:

• Exploiting, harming, or attempting to exploit or harm minors, including solicitation, creation, acquisition, or dissemination of child exploitative content.
• Generating or disseminating verifiably false information with the intent to harm others.
• Creating or distributing personal identifiable information that could be used to harm an individual.
• Harassing, abusing, threatening, stalking, or bullying individuals or groups.
• Producing non-consensual nudity or illegal pornographic content.
• Making fully automated decisions that adversely affect an individual’s legal rights or create binding obligations.
• Facilitating large-scale disinformation campaigns.

Accelerated Inference

Our accelerated inference approach leverages advanced optimization technology from WavespeedAI. This innovative fusion technique significantly reduces computational overhead and latency, enabling rapid image generation without compromising quality. The entire system is designed to efficiently handle large-scale inference tasks while ensuring that real-time applications achieve an optimal balance between speed and accuracy. For further details, please refer to the blog post.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v2/wavespeed-ai/wan-2.1/i2v-480p-ultra-fast" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "image": "https://d2g64w682n9w0w.cloudfront.net/media/images/1745079024013078406_QT6jKNPZ.png",
    "prompt": "A girl stands in a lively 17th-century market. She holds a red tomato, looks gently into the camera and smiles briefly. Then, she glances at the tomato in her hand, slowly sets it back into the basket, turns around gracefully, and walks away with her back to the camera. The market around her is rich with colorful vegetables, meats hanging above, and bustling townsfolk. Golden-hour painterly lighting, subtle facial expressions, smooth cinematic motion, ultra-realistic detail, Vermeer-inspired style",
    "negative_prompt": "",
    "size": "832*480",
    "num_inference_steps": 30,
    "duration": 5,
    "guidance_scale": 5,
    "flow_shift": 3,
    "seed": -1,
    "enable_safety_checker": true
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v2/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Query Parameters

Result Request Parameters

Result Response Parameters