BrowseAI

On-device AI chat for any web page via Chrome's Gemini Nano. No cloud calls, no tracking — fully private and offline-capable.

BrowseAI - Your AI Assistant for Every Page BrowseAI brings the power of Google's Gemini Nano AI directly to your browser, powered by an on-device model that gives you instant answers about any webpage you visit, with or without page context, without ever leaving your browser. This privacy-first model works completely offline: there's no data transmission or tracking, and all processing happens locally using Gemini Nano, so your conversations, browsing data, and prompts never leave your computer. Summarize content, ask follow-up questions, or switch to general assistant mode anytime. ⚠️ SYSTEM REQUIREMENTS Requires Google Chrome 151+ with built-in AI enabled. On first launch, download the Gemini Nano model (~2-4 GB) via the "Download Model" button and make sure you have sufficient disk space. ■ FEATURES (TOOLS) → Form Filler - Paste messy text like your resume or any document, and BrowseAI suggests values for the form fields on the page. How to fill forms with Form Filler: 1. Open the "Tools" dropdown and choose "Form Filler". 2. Paste unstructured data (for example, resume text), press Enter, and wait while the agent fills the form. 3. Review the suggestions, then either accept or dismiss each field one by one, or use "Accept All" or "Dismiss All". → Tables Extractor - Extracts all HTML tables from the current page using AI and prepares downloadable exports. How to use Tables Extractor: 1. Open the "Tools" dropdown and choose "Tables Extractor". 2. Press Enter with or without an additional prompt. 3. Use the download menu in the composer to export CSV, Excel, JSON, or Markdown files packed as one zip. ■ FEATURES (MIX) → AI Bookmarks - Automatically organizes the pages you visit into AI-generated categories, then lets you browse, expand, open, and delete bookmarks directly from the assistant panel. How to use AI Bookmarks: 1. Open Settings and keep "AI Bookmarks" enabled. 2. Visit pages normally; BrowseAI categorizes pages and stores bookmarks locally by category. 3. Open the chat header menu, choose "AI Bookmarks", expand categories, and open or delete entries. → Multimodal Support - Upload and analyze images and audio files alongside your text prompts. Ask questions about diagrams, charts, visual content, or use audio inputs to analyze recordings and transcribe speech for richer context understanding. → Capture Visible Area - Take a screenshot of the current visible webpage area and attach it directly to your prompt for image analysis. → Snip & Capture - Select any specific region of the webpage to capture and attach only that part for focused visual questions. → Selection Highlighting - See exactly which text you selected for context, displayed right in the chat window. → Context Mode Toggle - Switch between page-aware mode (uses full webpage for context) and general mode (pure AI assistant without page context) with one click. → Real-time Streaming - Watch AI responses appear in real-time as they're generated, giving you faster feedback and the ability to read along as the model thinks. → Context Window Usage — Shows how much of Gemini Nano's available AI context (working memory) is currently being used. ■ HOW TO GET STARTED Install BrowseAI and click the BrowseAI floating button to open the chat panel. Download the Gemini Nano AI model by clicking the Download Model button when prompted. Once the model is ready, you can start using the assistant. Select any text on a page and ask BrowseAI about it, or simply ask to summarize the current page. Switch between page context and general mode using the toggle at the bottom of the chat window. ■ PERFECT FOR Content creators analyzing and summarizing research materials. Students understanding complex topics and documentation. Developers quickly grasping code explanations and API references. Researchers comparing and extracting information from multiple pages. Professionals processing and summarizing lengthy reports. ⚠️ IMPORTANT DISCLAIMER BrowseAI is an assistive tool and includes custom tools that may generate or return incorrect, incomplete, misleading, or outdated outputs. Users are responsible for reviewing and verifying outputs before relying on them, especially for legal, financial, medical, security, or other high-impact decisions. BrowseAI must not be used for unlawful, harmful, deceptive, or unauthorized activities, including misuse of third-party content or services. The developer is not responsible for user misuse of the extension or for consequences resulting from mistakes in AI-generated output or custom tool output.