Quick Start

Getting Started

1. Create an Account

Visit WaveSpeedAI to create your account.

2. Start Creating

You can generate images in two ways:

API Integration

All WaveSpeed AI APIs require authentication. Please refer to our Authentication Guide for details on how to obtain and use your API key.

Choose the API service you need to get started:

• Flux Dev - Fast development version of the Flux model

• Flux Dev Lora - Flux model with Lora fine-tuning support

• Flux Schnell - High-speed Flux model with Flux Schnell

• Flux Schnell Lora - High-speed Flux model with Lora support

• WAN T2V 720p - High-resolution text-to-video model (720p)

• WAN T2V 720p Lora - High-resolution text-to-video with Lora (720p)

• WAN I2V 720p - High-resolution image-to-video model (720p)

• WAN I2V 720p Lora - High-resolution image-to-video with Lora (720p)

• WAN T2V 480p - Text-to-video conversion model (480p resolution)

• WAN T2V 480p Lora - Text-to-video model with Lora support (480p)

• WAN I2V 480p - Image-to-video conversion model (480p resolution)

• WAN I2V 480p Lora - Image-to-video model with Lora support (480p)

• Hunyuan T2V - Text-to-video model by Hunyuan

• Hunyuan I2V - Image-to-video model by Hunyuan

• Kling V1.6 I2v Pro - Text-to-video model by Kling

• Kling V1.6 I2v Standard - Text-to-video model by Kling with Lora support

• Kling V1.6 T2v Standard - Text-to-video model by Kling

• Minimax Video 01 - Text-to-video model by Minimax

Supported Programming Languages

Our APIs can be accessed via HTTP requests from any programming language, including:

• Python
• JavaScript/TypeScript
• Go
• Java
• PHP

For detailed API references and example code, please visit each model’s documentation page.

Web Interface

Click the model cards below to start creating directly in our web interface:

black-forest-labs/flux-dev (ultra fast)

For detailed API implementation and code examples, please refer to our Introduction page.

ComfyUI Integration

Integrate WaveSpeed AI’s high-performance image and video generation services directly into ComfyUI with our custom nodes.

Key Features

• Seamless API access within ComfyUI
• Multiple nodes for diverse AI tasks
• LoRA support for style customization

Quick Setup

1. Get your API key from WaveSpeed AI
2. Install ComfyUI-WaveSpeedAI-API custom node:

   cd ComfyUI/custom_nodes
   git clone https://github.com/WaveSpeedAI/wavespeed-comfyui.git
   cd wavespeed-comfyui
   pip install -r requirements.txt
3. Set your API key:
   • In the Client node, or
   • In config.ini (rename config.ini.tmp to config.ini and add your key)

For detailed node information and advanced usage, refer to our GitHub repository.

Available Models

Image Generation

1. FLUX-dev

   • 12B parameter high-performance model
   • Generation speed: < 2 seconds
   • Use case: General image generation

2. FLUX-lora

   • Fast image generation with LoRA support
   • Features: Personalization, style adaptation, brand identity customization

Video Generation

1. Hunyuan Video

   • High-quality video generation by Tencent
   • Features: High quality, realistic results
   • Generation time: ~2 minutes

2. WAN-2.1 T2V

   • Text-to-video generation
   • Diverse motion generation
   • Output: 5s 480/720p videos

3. WAN-2.1 I2V

   • Image-to-video generation
   • Dynamic video from static images
   • Maintains original image style