About This Tool

How to Use
  1. Select an AI model from the dropdown (smaller = faster download)
  2. Click 'Load Model' to download and cache the model in your browser
  3. Wait for the download to complete (700MB–2GB depending on model)
  4. Type your message in the input box and press Enter or click Send
  5. Optionally expand 'System Prompt' to customize the assistant's behavior
  6. Use 'Clear Chat' to start a fresh conversation
Common Use Cases
  • Private AI conversations with no data sent to any server
  • Offline AI assistance when internet is unavailable
  • Testing LLM behavior without API costs
  • Experimenting with different system prompts
  • Sensitive questions you don't want sent to cloud AI services
Tips & Tricks
  • The model is cached after the first download — subsequent visits load instantly
  • Smaller models (Llama 3.2 1B) are faster but less capable
  • Larger models (Phi 3.5 Mini) give better answers but need more RAM/VRAM
  • Requires Chrome 113+ or Edge 113+ with WebGPU support
  • Close other GPU-heavy tabs if the model fails to load

Related Tools

AI Code Explainer

Paste any code snippet and get AI-powered explanations, bug detection, or improvement suggestions. Runs entirely in your browser — no code is ever sent to a server.

Speech Studio

Convert text to speech with voice selection and speed controls, or transcribe speech to text in real-time. Uses built-in browser APIs — zero dependencies, total privacy.

Regex Query Helper & Explainer

Experiment with regular expressions using our interactive Regex Helper tool. Test your regex on sample text and get a breakdown of the syntax.

Random Password Generator

Generate secure, customizable passwords instantly. Choose length, character types, and complexity — perfect for creating strong credentials without any third-party service.