Aider + Ollama Local Setup

Configure Aider to use a local Ollama instance with the DeepSeek Coder model for completely private, offline AI pair programming. This setup keeps all code and interactions on your machine.

What This Does

This configuration sets up Aider to use your local Ollama server instead of cloud APIs, enabling:

Prerequisites

1. Install and run Ollama: `ollama serve`

2. Pull the DeepSeek Coder model: `ollama pull deepseek-coder:6.7b`

3. Verify Ollama is running on `http://localhost:11434`

Setup Instructions

1. **Create the Aider configuration file** in your project root or home directory:

Create `.aider.conf.yml`:

```yaml

AIDER configuration file

Use local Ollama instance with deepseek-coder model

model: ollama_chat/deepseek-coder:6.7b

openai-api-base: http://localhost:11434

Use solarized dark theme for code blocks

dark-mode: true

code-theme: solarized-dark

Keep everything local

stream: true

pretty: true

```

2. **Verify Ollama is running**:

```bash

curl http://localhost:11434/api/tags

```

3. **Start Aider** in your project directory:

```bash

aider

```

Aider will automatically detect and use the `.aider.conf.yml` configuration.

Configuration Breakdown

Usage Examples

Once configured, use Aider normally:

```bash

Start Aider with files

aider src/main.py

Add files during session

/add src/utils.py

Make changes

"Refactor the authentication function to use async/await"

Review changes

/diff

```

Switching Models

To use a different Ollama model, first pull it:

```bash

ollama pull codellama:13b

```

Then update `.aider.conf.yml`:

```yaml

model: ollama_chat/codellama:13b

```

Troubleshooting

**Aider can't connect to Ollama:**

**Model not found:**

**Slow responses:**

Notes