AI Chat: A Multilingual Multi-Model application

Streamlit

Python

LLM

Author

Tony D

Published

January 7, 2026

Introduction

In the rapidly evolving landscape of Artificial Intelligence, having a unified interface to interact with multiple models is invaluable. AI Chat is a multilingual, multi-model AI chat application built with Streamlit. It supports streaming responses, image inputs, concurrent model querying, web search, and comprehensive file processing.

Whether you’re comparing responses from different LLMs or generating stunning visuals, AI Chat provides a seamless and powerful experience.

AI summary paper:DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (https://arxiv.org/pdf/2501.12948)

✨ Key Features

🌐 Multilingual Support: Instant switching between English and Chinese interfaces.
🤖 Multi-Model Chat: Query multiple AI models simultaneously with a side-by-side comparison view.
🎨 Image Generation & Editing: Create and edit images using models like FLUX, Qwen Image, and Gemini.
📚 Prompt Bay: Access a library of 100+ searchable system prompts for specialized interactions.
🔍 Web Search: Integrated Tavily AI search for real-time information retrieval.
🖼️ Comprehensive File Support: Upload and process PDFs, Word docs, Excel, CSV, and images.
⚡ Streaming Responses: Real-time feedback from multiple models concurrently.

🚀 Quick Start

To get started with AI Chat locally, follow these steps:

Prerequisites

Python 3.7 or higher
pip package manager

Installation

Clone the repository:

git clone https://github.com/JCwinning/AI_Chat.git
cd AI_Chat

Install dependencies:
```
pip install -r requirements.txt
```

Configure API keys: Create a .env file in the project root and add your keys:

modelscope=your-modelscope-api-key
openrouter=your-openrouter-api-key
dashscope=your-dashscope-api-key
bigmodel=your-bigmodel-api-key
tavily_api_key=your-tavily-api-key

Run the application:
```
streamlit run app.py
```

🏗️ Architecture & Components

The project is structured for modularity and performance:

app.py: The main entry point using Streamlit for the UI and session management.
config.py: Centralized model definitions and provider settings.
search_providers.py: Handles web search integration with caching.
Multi-threading: Uses Python’s threading and Queue to handle parallel model requests and streaming.
File Processing: Leverages markitdown for document conversion and Pillow for image handling.

📖 Usage Tips

Side-by-Side Comparison

One of the most powerful features is the ability to select multiple models in the “🤖 Models” tab. When you send a message, the app queries all selected models in parallel and displays their responses side-by-side, making it easy to compare performance and accuracy.

Using the Prompt Bay

Don’t know how to start? Use the “📚 Prompt Bay” to find the perfect system prompt. You can search by category or keyword and apply it to your current session with a single click.

Advanced File Analysis

Upload a PDF or Excel file, and the app will automatically convert it to markdown or a table, including it in the conversation context. This allows you to ask the AI questions directly about your documents.

Check out the full source code on GitHub and start building your own AI-powered workflows!

--- title: "AI Chat: A Multilingual Multi-Model application" author: "Tony D" date: "2026-01-07" categories: [AI, Streamlit, Python, LLM] image: "images/featured.png" format: html: code-fold: true code-tools: true code-copy: true execute: warning: false --- # Introduction In the rapidly evolving landscape of Artificial Intelligence, having a unified interface to interact with multiple models is invaluable. [AI Chat](https://github.com/JCwinning/AI_Chat) is a multilingual, multi-model AI chat application built with Streamlit. It supports streaming responses, image inputs, concurrent model querying, web search, and comprehensive file processing. Whether you're comparing responses from different LLMs or generating stunning visuals, AI Chat provides a seamless and powerful experience. ::: {.panel-tabset} ## Text answer ![](images/1.png) ## Create image ![](images/2.png) ## summary upload document ![](images/4.png) AI summary paper:DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning (https://arxiv.org/pdf/2501.12948) ## pre define prompt ![](images/3.png) ::: # ✨ Key Features - **🌐 Multilingual Support**: Instant switching between English and Chinese interfaces. - **🤖 Multi-Model Chat**: Query multiple AI models simultaneously with a side-by-side comparison view. - **🎨 Image Generation & Editing**: Create and edit images using models like FLUX, Qwen Image, and Gemini. - **📚 Prompt Bay**: Access a library of 100+ searchable system prompts for specialized interactions. - **🔍 Web Search**: Integrated Tavily AI search for real-time information retrieval. - **🖼️ Comprehensive File Support**: Upload and process PDFs, Word docs, Excel, CSV, and images. - **⚡ Streaming Responses**: Real-time feedback from multiple models concurrently. # 🚀 Quick Start To get started with AI Chat locally, follow these steps: ### Prerequisites - Python 3.7 or higher - pip package manager ### Installation 1. **Clone the repository**: ```bash git clone https://github.com/JCwinning/AI_Chat.git cd AI_Chat ``` 2. **Install dependencies**: ```bash pip install -r requirements.txt ``` 3. **Configure API keys**: Create a `.env` file in the project root and add your keys: ```env modelscope=your-modelscope-api-key openrouter=your-openrouter-api-key dashscope=your-dashscope-api-key bigmodel=your-bigmodel-api-key tavily_api_key=your-tavily-api-key ``` 4. **Run the application**: ```bash streamlit run app.py ``` # 🏗️ Architecture & Components The project is structured for modularity and performance: - **`app.py`**: The main entry point using Streamlit for the UI and session management. - **`config.py`**: Centralized model definitions and provider settings. - **`search_providers.py`**: Handles web search integration with caching. - **Multi-threading**: Uses Python's `threading` and `Queue` to handle parallel model requests and streaming. - **File Processing**: Leverages `markitdown` for document conversion and `Pillow` for image handling. # 📖 Usage Tips ### Side-by-Side Comparison One of the most powerful features is the ability to select multiple models in the "🤖 Models" tab. When you send a message, the app queries all selected models in parallel and displays their responses side-by-side, making it easy to compare performance and accuracy. ### Using the Prompt Bay Don't know how to start? Use the "📚 Prompt Bay" to find the perfect system prompt. You can search by category or keyword and apply it to your current session with a single click. ### Advanced File Analysis Upload a PDF or Excel file, and the app will automatically convert it to markdown or a table, including it in the conversation context. This allows you to ask the AI questions directly about your documents. --- *Check out the full source code on [GitHub](https://github.com/JCwinning/AI_Chat) and start building your own AI-powered workflows!*