A modern multimodal AI chat UI that supports text, voice, and image inputs, backed by Ollama for running LLMs locally. Built using React, Gradio, FastAPI (optional), and integrated with multiple LLMs (DeepSeek, Mistral, LLaMA3, etc.).

• 🔤 Text input: Ask questions or give instructions in plain text.
• 🎤 Voice input: Upload .wav audio or speak directly (speech-to-text using Google Speech API).
• 🖼️ Image input: Upload image files for OCR using easyocr.
• 🤖 Model selection: Choose from available Ollama models dynamically.
• 🗃️ Conversation history: Stored per-session with conversation_id.
• 💾 Export chat: Download chat as JSON.
• 🚀 Streaming replies: Simulated (can be expanded with websocket).
• 🧪 Model introspection: Auto-fetch from /api/tags and fall back to defaults.

1. Install dependencies:

      python -m venv venv
      venv\Scripts\activate.bat
      pip install gradio easyocr speechrecognition ollama
      npm i

2. Run App:

      npm start