Wan 2.1 V2V 720p Ultra Fast

Turbo-charged inference for Wan 2.1 14B. Unleashing high-res 720p video-to-video prowess with cutting-edge suite of video foundation models

Features

Built on the Wan 2.1 series, Wan 2.1 V2V model enables controlled and expressive transformations—ideal for style transfer, character editing, and storyboarding—turning ordinary clips into creative visual narratives.This model combines the personalization capabilities of wavespeed-ai with ultra-fast image generation, delivering high-quality outputs in under 2 seconds.

Key Features:

• High-quality output: Wan 2.1 supports a wide range of video generation tasks and is capable of generating high-quality video content suitable for a variety of application scenarios.
• Complex Motion Simulation: It specializes in generating realistic videos with complex body movements, rotations, dynamic scene transitions, and smooth camera movements.
• Controlled editing and style shifting: Wan 2.1 provides a generic editing model that allows precise editing based on image or video references, ideal for style shifting, character editing, and storyboarding.

ComfyUI

wan-2.1/v2v-720p-ultra-fast video to video is available on ComfyUI, providing local inference capabilities through a node-based workflow, ensuring flexible and efficient image generation on your system.

Use Cases

• Style Transfer: Converts real-world footage into styles like animation, claymation, or pixel art, enabling creative content production and visual effects creation.
• Motion Transfer and Expansion: Transfers the motion structure from one video to another character, allowing rapid generation of new actions or shots in game development and virtual filmmaking.
• Creative Content Generation: Produces personalized video content for individuals or brands, enhancing their impact on social media platforms.

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/v2v-720p-ultra-fast" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "video": "https://d2g64w682n9w0w.cloudfront.net/media/ec44bbf6abac4c25998dd2c4af1a46a7/videos/1747054235924170377_ljxByuqm.mp4",
    "prompt": "A woman races up the staircase as rain pours down outside at night.",
    "negative_prompt": "",
    "num_inference_steps": 30,
    "duration": 5,
    "strength": 0.9,
    "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