QwenPulse-AI

A decoupled, full-stack local generative AI application featuring a highly interactive chat interface powered by local inference.

This project serves as an architectural implementation blueprint, mirroring enterprise-grade AI deployment patterns while remaining lightweight and fully local.

---

🏗️ System Architecture

The application implements a decoupled architectural pattern, separating presentation, routing, inference, and compute layers.

+---------------------------+
|  Presentation Layer       |
|  (Streamlit Frontend)     |
+---------------------------+
            |
            | Asynchronous HTTP Requests
            v
+---------------------------+
|  REST API Gateway         |
|  (OpenAI-Compatible Spec) |
+---------------------------+
            |
            | Internal Model Routing
            v
+---------------------------+
|  Inference Serving Engine |
|  (Ollama Orchestration)   |
+---------------------------+
            |
            | Hardware Acceleration
            v
+---------------------------+
|  Compute Environment      |
|  (Apple Silicon M4 Core)  |
+---------------------------+

Components

1. Presentation Layer (Streamlit)
   A reactive Python interface that manages chat interactions and real-time token streaming.

2. REST API Gateway
   Standardized routing utilizing the OpenAI-compatible API specification.

3. Inference Serving Engine (Ollama)
   A lightweight background service managing local model execution and orchestration.

4. Compute Layer (Apple Silicon)
   Optimized for unified memory architecture and Metal acceleration.

---

⚡ Tech Stack

• Frontend UI: Streamlit (Reactive Python-based UI with active token streaming)
• Inference Engine: Ollama CLI (Local model hosting and request orchestration)
• Model Architecture: Qwen 3.5 9B

---

🚀 Getting Started & Deployment

1. Prerequisites & Engine Initialization

Ensure you have Ollama installed on your system.

Pull and initialize the model:

ollama run qwen3.5:9b

---

2. Application Installation & Dependencies

Clone the repository:

git clone https://github.com/nerdsuryansh/QwenPulse-AI.git

Navigate to the project directory:

cd QwenPulse-AI

Create and activate a virtual environment:

python3 -m venv venv
source venv/bin/activate

Install required dependencies:

pip install -r requirements.txt

---

3. Launching the Local Frontend Gateway

Start the Streamlit application:

streamlit run src/app.py

Once launched, open:

http://localhost:8501

in your browser to interact with the application.

---

📁 Project Structure

QwenPulse-AI/
├── .gitignore          # Prevents tracking of environments and cache
├── README.md           # System documentation and deployment guide
├── requirements.txt    # Project dependency configuration
└── src/
    └── app.py          # Streamlit application and API connection logic

---

🛡️ License

This project is open-source and available under the MIT License.