Olla - Obsidian Ollama Integration

Olla is an Obsidian plugin that integrates Ollama into your note-taking workflow. It allows you to chat with local AI models directly within Obsidian.

Features

• Dockable Chat Panel: A dedicated panel for chatting with Ollama models
• Ribbon Toggle: Quick access via Obsidian's ribbon bar
• Model Management: View, download, and switch between available Ollama models
• Markdown Rendering: Ollama responses are rendered with full markdown support
• Persistent History: Chat history is saved per vault
• Connection Status: Visual indicator of Ollama server connection

Installation

Prerequisites

1. Ollama: You must have Ollama installed and running on your system.
2. Obsidian: Version 1.0.0 or higher.

Manual Installation

1. Download the latest release from the GitHub repository
2. Extract the plugin files to your Obsidian vault's plugins folder: VaultFolder/.obsidian/plugins/olla/
3. In Obsidian, go to Settings > Community plugins and enable "Olla"

Development Installation

1. Clone this repository into your Obsidian vault's plugins folder
2. Run npm install to install dependencies
3. Run npm run build to compile the plugin
4. Enable the plugin in Obsidian under Settings > Community plugins

Usage

Starting a Chat

1. Click the Olla icon in the ribbon bar (or use the command palette)
2. The chat panel will open on the right side of your workspace
3. Select a model from the dropdown menu
4. Type your message and press Enter to send

Managing Models

1. Open Olla settings from the Obsidian settings panel
2. View all available models that are currently downloaded
3. To download a new model:
   • Enter the model name (e.g., llama2, mistral) in the "Model Name" field
   • Click "Download"
   • The model will be downloaded and appear in the available models list

Changing Server URL

By default, Olla connects to http://localhost:11434. If your Ollama server is running on a different URL:

1. Open Olla settings
2. Update the "Ollama Server URL" field
3. Save the settings

Configuration

The plugin automatically saves your settings including:

• Ollama server URL
• Selected model
• Chat history
• Available models

Troubleshooting

Ollama is not connected

• Make sure Ollama is running: ollama serve
• Check that the server URL in settings is correct
• Verify that the Ollama port (default: 11434) is accessible

Models not appearing

• Ensure you have downloaded models using Ollama: ollama pull llama2
• Restart the Ollama server if you recently downloaded models
• Check the connection status in the chat panel

Plugin not loading

• Make sure you've run npm run build after making changes
• Check the Obsidian console for errors (Ctrl+Shift+I > Console)
• Ensure you're using a compatible version of Obsidian

Development

Project Structure

• main.ts: Main plugin file with all functionality
• manifest.json: Plugin metadata
• package.json: Project configuration and dependencies
• tsconfig.json: TypeScript configuration

Building

npm run build

For development with automatic rebuilding:

npm run dev

Architecture

The plugin uses:

• Obsidian API for plugin integration
• React for UI components
• TypeScript for type safety
• Ollama REST API for model interaction

Contributing

Contributions are welcome! Please open an issue or pull request on GitHub.

License

MIT