ChatAI🧠 Local Multimodal AI Assistant Sidebar 作者： ChatAI🧠

Open source local multimodal AI assistant for firefox featuring vision, voice, RAG, with Ollama/OpenAI compatibility.

Bring the full force of cutting-edge artificial intelligence right into your Firefox sidebar—entirely offline, entirely secure.

ChatAI🧠 bridges the gap between your browser and local intelligence engines, empowering you to process documents, parse images, write code, and converse with local LLMs without leaking a single byte of data to third-party cloud servers.

🌟 Why ChatAI🧠?

Zero Cloud Reliance: Connect directly to your local Ollama backend or any OpenAI-compatible API. Your conversations, prompts, and files never leave your machine.

Multimodal Vision: Drag, drop, and analyze screenshots, UI designs, or photos instantly using local vision-language models.

Contextual RAG: Feed PDFs, documentation, and web snippets to your models to instantly query your private knowledge base.

Developer-Ready Workspace: Built-in syntax highlighting, instant code-block extraction, and prompt templates designed for high-velocity coding.

Voice-Enabled: Speak your prompts fluidly for hands-free browsing and interaction.


⚙ Quick Start Guide

To unlock ChatAI🧠's full vision and text capabilities, set up a local runtime in minutes:

Get Ollama: Download and install the runtime for your system from https://ollama.com/download.

Pull a Multimodal Model: Fire up your terminal and download a powerful local model to handle both text and images:

ollama pull minicpm-v4.6 (good for cpu only system)


or

ollama pull gemma4


or
ollama pull qwen3.6

Launch & Sync: Open ChatAI🧠 in your Firefox sidebar, point it to your local Ollama port, and experience true private browsing intelligence.