Frequently Asked Questions

LumozAI currently supports various types of models, including:

• Stable Diffusion LoRA models for image generation
• Flux LoRA models with enhanced performance
• Other compatible AI models for various use cases

We're constantly expanding our supported model types.

Usage rights depend on the specific model's license. Each model on our platform displays its license information, which determines how you can use it. Some models permit commercial use, while others are for personal or research purposes only.

Always check the license details before using a model for commercial projects.

We have a multi-step process for model verification:

• Automated checks for compatibility and performance
• Review of model documentation and intended use
• Testing for basic functionality and compliance with our terms

Additionally, our user rating system helps identify high-quality models.

Yes! We welcome model requests. You can submit your suggestions through our contact page or by emailing us directly. While we can't guarantee that every requested model will be added, we value user input for expanding our library.

Deploying your model involves three simple steps:

1. Select the model type (SD LoRA, Flux LoRA, etc.)
2. Configure your model (name, URL, sample prompt, privacy settings)
3. Submit for deployment

Our system will handle the rest and notify you once deployment is complete.

To deploy a model, you'll need:

• A model hosted on HuggingFace, CivitAI, or similar platforms
• Model name and description
• Sample prompt (optional, will be generated automatically if not provided)
• Sufficient account balance for deployment costs

Your model must also comply with our terms and conditions.

Yes, you have two privacy options when deploying models:

• Public: Your model will be visible to all users and listed in our model directory
• Private: Your model will only be accessible to you and users you explicitly grant access to

You can select your preferred privacy setting during the deployment process.

Deployment typically takes between 5-15 minutes, depending on model size and complexity. During this time, our system downloads the model, sets up dependencies, configures parameters, and prepares it for use.

You'll receive a notification when your model is ready to use.

LumozAI operates on a pay-per-usage model. Here's how it works:

• You load balance to your account using our Stripe payment gateway
• As you use models or deploy your own, costs are deducted from your balance
• You only pay for what you use
• You can add more funds at any time

This approach gives you flexibility without subscription commitments.

Through our Stripe integration, we accept:

• Major credit cards (Visa, Mastercard, American Express, etc.)
• Debit cards
• Apple Pay and Google Pay (where available)

All transactions are secure and encrypted.

No, balance loaded to your account is non-refundable. We recommend loading only the amount you plan to use in the near future.

Your balance never expires, so you can use it at any time for our services.

Costs vary depending on several factors:

• The specific model being used
• Computing resources required
• Duration and frequency of use

Each model displays its usage cost before you run it. For deployed models, costs include both the initial deployment and subsequent usage.

We use high-performance cloud infrastructure optimized for AI workloads, including:

• NVIDIA GPUs for accelerated computing
• Scalable compute resources that adjust based on demand
• High-bandwidth networking for fast data transfer

Our infrastructure is designed to provide optimal performance for various AI model types.

Yes, we offer API access to deployed models. This allows you to integrate LumozAI capabilities directly into your applications. API keys can be generated from your account dashboard, and comprehensive API documentation is available in our developer resources section.

Supported output formats depend on the model type:

• Image generation models: PNG, JPEG, WebP
• Text models: Plain text, JSON, Markdown
• Other formats depending on specific model capabilities

Output format options are displayed when using each model.

We take data security seriously:

• All data in transit is encrypted using industry-standard TLS
• User authentication employs secure practices
• We implement strict access controls to protect your models and data
• Regular security audits are performed on our infrastructure

For more details, please refer to our Privacy Policy and Terms of Service.