Chat with Code using Qwen3-Coder

Enhance your experience with GitHub repositories through a natural language interface. We are developing a Streamlit app that enables users to communicate with code using the Qwen3-Coder model. This app offers a user-friendly interface for querying code and receiving responses, along with the additional advantage of validating those responses using Cleanlab Codex.

We use:

• Llama_Index for orchestration
• Milvus to self-host a VectorDB
• Cleanlab codex to validate the response
• OpenRouterAI to access Alibaba's Qwen3-Coder

Set Up

Follow these steps one by one:

Setup Milvus VectorDB

Milvus provides an installation script to install it as a docker container.

To install Milvus in Docker, you can use the following command:

curl -sfL https://raw.githubusercontent.com/milvus-io/milvus/master/scripts/standalone_embed.sh -o standalone_embed.sh

bash standalone_embed.sh start

Install Dependencies

uv sync

Run the Notebook

You can run the notebook.ipynb file to test the functionality of the code in a Jupyter Notebook environment. This notebook will guide you through the process of querying code and validating responses.

Run the Application

To run the Streamlit app, use the following command:

streamlit run app.py

Open your browser and navigate to http://localhost:8501 to access the app.

📬 Stay Updated with Our Newsletter!

Get a FREE Data Science eBook 📖 with 150+ essential lessons in Data Science when you subscribe to our newsletter! Stay in the loop with the latest tutorials, insights, and exclusive resources. Subscribe now!

Contribution