A Step-by-Step Tutorial for Installing and Using aider AI-powered coding tool

aider.chat is an open-source tool designed for pair programming with LLMs, enabling you to edit code in your local Git repository. Aider works best with GPT-4o and Claude-3.5 Sonnet and can connect to almost any LLM. This tutorial walks you through the process of installing aider and using it in a practical scenario to enhance your productivity.

Prerequisites

Before we get started, ensure you have the following:

Operating System: Linux (preferred).
Python: Python 3.8 or higher installed on your system.
Access to Terminal: Basic familiarity with using the terminal/command line.

Environment Note: This tutorial was performed on an Ubuntu WSL (Windows Subsystem for Linux) setup running on a Windows 11 personal computer. The steps should be similar for other Linux-based environments.

Step 1: Installing aider.chat using pipx

To install aider.chat, it is recommended to use pipx, which allows you to install and run Python applications in isolated environments. This prevents dependency conflicts between different Python projects on your system. Follow the instructions on the aider.chat pipx installation page.

To install aider.chat using pipx:

Install pipx (if not already installed).
Install aider.chat with pipx.

After installation, you can run aider.chat directly from the command line.

Step 2: Integrating aider.chat with an LLM Model

In this step, we will integrate aider.chat with the Anthropic Claude-3.5-Sonnet LLM model using the OpenRouter provider.

To learn more about the model and the provider, you can refer to the following links:

Setting Up the Integration

Obtain API Keys: Sign up and get your API key from OpenRouter.

Step 3: Running aider Tool from the Command Line

In this step, we will navigate to the root directory of a Git repository that contains a Python project and run the aider tool from the command line. For this tutorial, I have used a personal Python project that includes various Python tools used for working on my blog.

Navigate to the Git Repository: Move to the root directory of your project.
Set the Environment Variable: Ensure the environment variable OPENROUTER_API_KEY is set with your API key.
Run aider with the LLM Model: Use the command line to start aider with the Claude-3.5-Sonnet model via OpenRouter:

aider --model openrouter/anthropic/claude-3.5-sonnet

Step 4: Adding a Python File to the Chat

For the following practical cases, we will be using a simple Python tool in my project. This tool automates the translation of blog posts from English to other languages using Camel AI agents integrated with LLM API providers.

Add a Python File: Add a Python file to the aider chat session for interactive collaboration with this aider command.

/add CAMEL_translate_tool_mistral.py

And basically, you are ready to go with these practical cases to show aider’s capabilities as your AI-powered coding assistant.

Practical Case 1: Adding docstrings to Python methods

Instruct aider to add docstrings: Use aider to generate detailed docstring documentation for methods in your Python script.

Results: The results were positive. Aider successfully added clear and accurate docstrings, improving the script's readability and maintainability.

API LLM Tokens Used and Cost: These are shown in the above picture.

Practical Case 2: Adding Support for GROQ modelfactory

Now a more difficult task. My script is integrated with two LLM API providers: Mistral and Gemini. In this case, we instructed aider to add support for the GROQ provider, using for inference the Groq Llama 3 8B model. Aider updated the script to include the necessary imports, model initialization, and adjustments for compatibility.

Adding GROQ Support Figure 1

Adding GROQ Support Figure 2

Adding GROQ Support Figure 3

Results

While the integration was mostly successful, two issues were identified:

Incorrect Model Setting: Aider initially configured the model incorrectly.
Incorrect GROQ API URL: Aider failed to set the correct URL for the GROQ API, resulting in integration issues.

I instructed aider to set the API URL, but the results were unsatisfactory. Consequently, I reverted the changes using the aider /undo command.

Model Setting Issue Figure

Finally, I manually updated the script to ensure both the model configuration and the API URL were correctly set.

Manual Edit for URL Figure

After addressing these issues manually, testing confirmed that the model and API were successfully integrated and functioning as expected.

Practical Case 3: Modifying the commit method for branch creation

Update the Commit Method: Instruct aider to modify the commit logic in my script to create the target branch if it doesn’t exist.

Results: The target branch was successfully created when it did not exist, and changes were committed without issues, confirming aider’s capabilities.

Conclusions

User-Friendly: The steps to install and use aider.chat are simple and straightforward, making it accessible to anyone with basic command-line experience.
Positive Results: Overall, the tool provided positive results, effectively achieving the modifications we aimed for.
Basic Use Case: This tutorial demonstrated a basic use case, modifying just one Python script. More complex use cases may be explored in future posts.
Integration with Git: Aider.chat's seamless integration with Git made version control and branch management easy, ensuring efficient tracking of changes.
Cost Considerations: When using the LLM API, it's important to factor in the associated costs, particularly for extensive or production-level use.

Check Out Other Tutorials on My Blog

If you found this tutorial helpful, you might enjoy these as well:

Enjoyed this post? Found it helpful? Feel free to leave a comment below to share your thoughts or ask questions. A GitHub account is required to join the discussion.