Instant Character

InstantCharacter creates high-quality, consistent characters from text prompts, supporting diverse poses, styles, and appearances with strong identity control.

Features

InstantCharacter is a cutting-edge model designed for open-domain character personalization. It addresses the common limitations of traditional models in terms of generalization ability and image quality. By leveraging a scalable Diffusion Transformer architecture, combined with adaptable adapter modules and a massive character dataset, InstantCharacter delivers high-quality, highly customizable character imagery across diverse scenarios.

Key Features

• Open-Domain Personalization：InstantCharacter excels in generating a wide variety of character appearances, poses, and styles, maintaining high fidelity while allowing extensive personalization — perfect for a wide range of creative and professional applications.
• Scalable Adapter Modules：The model integrates stacked Transformer encoders as adapters that specialize in handling open-domain character features. These adapters interact seamlessly with the latent space of the modern Diffusion Transformer, significantly enhancing the diversity and consistency of generated results.
• Large-Scale Character Dataset Training：To empower InstantCharacter’s capabilities, Tencent AI Lab built a massive character dataset containing tens of millions of samples, covering multi-view character images and paired text-image examples. This robust dataset enables superior identity consistency and strong text-based editability.

ComfyUI

InstantCharacter is also available on ComfyUI, providing local inference capabilities through a node-based workflow, ensuring flexible and efficient image generation on your system.

Use Cases

• Game Development Rapidly create a wide variety of high-fidelity character assets, accelerating production pipelines.
• Virtual Reality Generate immersive, personalized avatars to enhance user experiences.
• Digital Content Creation Produce diverse character designs for animations, comics, and more, empowering creative expression.
• Social Media Create personalized profile pictures and stickers that reflect individual styles and preferences.

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/instant-character" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "prompt": "A futuristic android girl with glowing blue eyes, wearing green suit, standing next to a holographic sign displaying \"I Love WaveSpeedAI\", neon-lit cyberpunk city background, sleek and high-tech vibe, Sci-fi movie style",
    "image": "https://d2g64w682n9w0w.cloudfront.net/media/images/1745853771145444122_WW40XTQM.jpg",
    "size": "1024*1024",
    "negative_prompt": "",
    "guidance_scale": 3.5,
    "num_inference_steps": 28,
    "num_images": 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