Local Reasoning Chatbot using Ollama and qwen3 Model

This project sets up a chatbot locally using Ollama and the qwen3 model.

Prerequisites

• Operating System: Windows, macOS, or Linux
• Ollama: Installed locally. Download Ollama
• Terminal or command line access
• Python (Optional): For frontend/backend integration

Model Used

• Model: qwen3:0.6b
• This is a compact, open-source large language model from Alibaba, suited for local inference.

Setup Instructions

1. Install Ollama

Download and install Ollama from the official site:
👉 https://ollama.com/download

Once installed, verify it's working:

ollama --version

2. Pull the qwen3 Model

Use the following command to download the model:

ollama pull qwen3:0.6b

You can explore other versions with:

ollama run qwen3

3. Run the Chatbot Locally

To start a chatbot session with the qwen3 model:

ollama run qwen3:0.6b

You’ll enter a REPL interface where you can chat directly with the model.

4. Optional: API Integration

Ollama exposes a local HTTP API at http://localhost:11434.

You can make requests like this:

curl http://localhost:11434/api/generate -d '{
  "model": "qwen3:0.6b",
  "prompt": "Hello!",
  "stream": false
}'

5. Use with Frontend (e.g., Streamlit, Flask)

Run with:

streamlit run chat.py

---

Troubleshooting

• Port Conflict: Ensure no other service is using port 11434.
• Model Not Found: If qwen3:0.6b isn't available, run ollama list to check available models.

License

This project is for personal/local use. Check qwen3 and Ollama licenses for distribution or commercial usage.