Wan 2.1 14B VACE

VACE is an all-in-one model designed for video creation and editing. It encompasses various tasks, including reference-to-video generation (R2V), video-to-video editing (V2V), and masked video-to-video editing (MV2V), allowing users to compose these tasks freely. This functionality enables users to explore diverse possibilities and streamlines their workflows effectively, offering a range of capabilities, such as Move-Anything, Swap-Anything, Reference-Anything, Expand-Anything, Animate-Anything, and more.

Features

wan-2.1-14b-vace provides solutions for various tasks, including reference-to-video generation (R2V) and video-to-video editing (V2V), allowing creators to freely combine these capabilities to achieve complex tasks.

Key Features

• All-in-One Video Processing: Supports multiple tasks including Image-to-Video (I2V),reference-to-video(R2V) and Video Editing, allowing for comprehensive video creation and editing workflows.
• High Performance: The model comes in 14B parameters and offers higher fidelity outputs.
• SOTA Performance: Wan2.1 consistently outperforms existing open-source models and state-of-the-art commercial solutions across multiple benchmarks.
• Unified Multi-task Capability: Extensive experiments show that VACE performs on par with task-specific models across various subtasks, while also enabling diverse applications through flexible task combinations.

ComfyUI

wan-2.1-14b-vace 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: Convert real-world videos into distinct styles such as animation, claymation, or pixel art, enabling unique visual storytelling for creators, filmmakers, and advertisers.
• Motion Transfer & Expansion: Apply motion from a source video to a new subject or character, allowing fast prototyping of new animations, cinematic shots, or in-game sequences.
• Virtual Try-On & Product Customization: Use masked video editing to seamlessly change clothing, backgrounds, or products within a video without reshooting—perfect for e-commerce and digital showrooms.
• Game & Character Animation: Create or edit character actions, environmental interactions, or cinematic sequences, streamlining game development and virtual production pipelines.

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/v3/wavespeed-ai/wan-2.1-14b-vace" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "images": [
        "https://d2g64w682n9w0w.cloudfront.net/media/f82c31a9020c49459ec9e57f26d0a22f/images/1747909398229595025_hWgc952Y.png",
        "https://d2g64w682n9w0w.cloudfront.net/media/dd703f8dc97646268d372cdd689fdccc/images/1747877882344729896_SU9630WS.png"
    ],
    "video": "",
    "task": "depth",
    "prompt": "The elegant lady carefully selects bags in the boutique, and she shows the charm of a mature woman in a black slim dress with a pearl necklace, as well as her pretty face. Holding a vintage-inspired blue leather half-moon handbag, she is carefully observing its craftsmanship and texture. The interior of the store is a haven of sophistication and luxury. Soft, ambient lighting casts a warm glow over the polished wooden floors",
    "negative_prompt": "",
    "size": "832*480",
    "num_inference_steps": 30,
    "enable_fast_mode": false,
    "guidance_scale": 5,
    "flow_shift": 16,
    "context_scale": 1,
    "seed": -1,
    "enable_safety_checker": true
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/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