Documentation

Welcome to InfiNet AI Documentation. This guide will help you get started with our uncensored AI platform and make the most of its powerful features.

Getting Started

InfiNet AI is a web-based platform that provides unrestricted access to advanced AI models. Simply navigate to the chat interface and start conversing with our AI models.

Note: All your conversations and generated content are fully owned by you, keeping your data completely private.

Model Selection

InfiNet AI offers two specialized models, each optimized for different use cases:

InfiNet-Coder

Switch to Coder mode (toggle OFF) for programming, code generation, debugging, and security research tasks. This model uses the Qwen-3 Q4_K_M-GGUF 35B architecture, specifically fine-tuned for technical development work.

Code Generation Generate code in any programming language with high accuracy and best practices.
Debugging Get detailed explanations of errors and suggested fixes for your code.
Security Research Analyze vulnerabilities, penetration testing, and ethical hacking scenarios.
Code Review Receive comprehensive code reviews and optimization suggestions.

InfiNet-Thinker

Switch to Thinker mode (toggle ON) for general reasoning, creative writing, problem-solving, using the Qwen-3 32B architecture. This model excels at complex thought processes and creative tasks.

Creative Writing Generate stories, articles, scripts, and creative content without restrictions.
Problem Solving Get detailed analysis and solutions for complex problems across all domains.
Reasoning Engage in deep reasoning, logical analysis, and strategic thinking.

InfiNet-Realistic-Vision-v60

Use the image generation feature for creating high-quality images from text descriptions (Text-to-Image) or modifying existing images based on your prompts (Image-to-Image). This model uses the Realistic-Vision-v60 architecture, optimized for photorealistic image generation and advanced image manipulation.

Text-to-Image Generate photorealistic images from text descriptions with high detail and accuracy.
Image-to-Image Transform and modify existing images based on your prompts while preserving key elements.
Photorealistic Quality Create stunning, realistic images suitable for professional use and creative projects.
Advanced Control Fine-tune image generation with detailed prompts and image reference inputs.

How to Switch Models

To switch between Coder and Thinker models:

  1. Locate the model toggle switch in the chat input area (left side)
  2. Click the toggle switch to change models
  3. The label will update to show "Coder" (OFF) or "Thinker" (ON)
  4. Your conversation will use the selected model for responses
Coder Mode: Toggle OFF → For coding, hacking, development
Thinker Mode: Toggle ON → For reasoning, creative tasks

Using the Chat Interface

Basic Chat

Type your text message or use voice recording in the chat input field and press Enter or click the send button. The AI will respond based on your selected model.

Keyboard Shortcuts

  • Enter - Send message
  • Shift + Enter - New line in message

Image Upload

You can upload images for image-to-image processing:

  1. Click the image icon button in the chat input area
  2. Select an image file from your device
  3. Describe what changes you want to make
  4. The AI will process your request using image generation capabilities

Conversation Management

Creating New Conversations

Click the "+ New Chat" button in the sidebar to start a fresh conversation. Each conversation is saved locally in your browser.

Conversation History

All your conversations are displayed in the sidebar. Click on any conversation to continue where you left off. Conversations are stored locally in your browser's localStorage.

Conversation Privacy

While conversations are stored locally in your browser, remember that no data from any generated content is stored on our servers except user IDs. For maximum privacy, use private/incognito browsing mode.

Performance & Speed

Our models use Q4_K_M quantization with GGUF format for optimal performance:

  • Ultra-Low Latency: Average response time under 500ms
  • Efficient Memory: Optimized quantization reduces memory usage by 75%
  • High Quality: Q4_K_M maintains 95%+ accuracy compared to full precision models
  • Fast Inference: GGUF format enables rapid token generation

Privacy & Security

Data Handling

InfiNet AI is designed with privacy as a core principle:

  • No conversation logging on servers
  • Local storage only in your browser
  • No third-party tracking
  • End-to-end encryption for sensitive operations

Best Practices

Important: For maximum security when working with sensitive information, consider using private browsing mode and clearing your browser data after sessions.

Token Usage

Free accounts receive 5,000 tokens to get started. Tokens are consumed based on:

  • Input message length
  • Model response length
  • Image processing requests

Check your token balance in the header. Upgrade to a paid plan for higher limits and faster speeds.

Troubleshooting

Slow Responses

If responses are slow, try:

  • Reducing the length of your input message
  • Closing other browser tabs to free up resources
  • Checking your internet connection

Model Not Switching

If the model toggle doesn't work:

  • Refresh the page
  • Clear your browser cache
  • Ensure JavaScript is enabled

Conversations Not Saving

If conversations aren't appearing:

  • Check if localStorage is enabled in your browser
  • Clear cookies and localStorage, then try again
  • Use a modern browser (Chrome, Firefox, Safari, Edge)

Advanced Usage

Code Generation Tips

For best results with InfiNet-Coder:

  • Be specific about programming language and framework
  • Include error messages or relevant context
  • Request explanations for complex code sections
  • Ask for code reviews and optimization suggestions

Image Generation Tips

When using image generation features:

  • Provide detailed descriptions for better results
  • Specify style, mood, and composition preferences
  • Use image-to-image for specific modifications
  • Experiment with different prompts to refine outputs

API Integration

For developers looking to integrate InfiNet AI into their applications, API access is available through our backend. Contact support for API credentials and documentation.

Note: API access requires a Pro+ or Enterprise plan. See our Pricing page for details.

Support

Need help? Contact us at [email protected] or visit our website at infinet.services for more information.

Version Information

Current platform version: 1.0.0

  • InfiNet-Coder: Qwen-3 Q4_K_M-GGUF 35B
  • InfiNet-Thinker: Qwen-3 Q4_K_M-GGUF 32B
  • InfiNet Realistic-Vision-v60: Realistic-Vision-v60 (Text-to-Image & Image-to-Image)
  • Platform: Web-based, no installation required