FAQ

FAQ

General

What is LocalAI Code Editor?

A free, open-source, self-hosted AI code editor. It connects to your local LLM (Ollama, LM Studio, llama.cpp, or vLLM) to provide AI-powered coding assistance without sending data to the cloud.

Is it really free?

Yes. It's MIT licensed and completely free to use.

Does it send data to the cloud?

No. Everything runs locally. Your code and conversations never leave your machine.

What LLMs are supported?

Any OpenAI-compatible API:

• Ollama (recommended)
• LM Studio
• llama.cpp server
• vLLM
• Any other OpenAI-compatible endpoint

Setup

How do I install it?

Option 1: Docker (easiest)

docker compose up

Option 2: From source

git clone https://github.com/wildfirebill-ai/localai-code-editor.git
cd localai-code-editor
pnpm install
pnpm dev:desktop

Option 3: Download Download the latest release for your platform from GitHub Releases.

How do I connect to Ollama?

1. Install Ollama: curl -fsSL https://ollama.com/install.sh | sh
2. Pull a model: ollama pull codellama:7b
3. Start the server: ollama serve
4. In the editor, go to Settings → LLM Providers
5. Click "+ Add provider" → Select "Ollama" preset

How do I use MCP servers?

1. Go to Settings → MCP Servers → Discover
2. Browse or search for a server
3. Click "Install"
4. Configure any required environment variables
5. The server's tools appear in Connected Tools

Features

What are Skills?

Skills are markdown instructions that teach the agent how to perform specific tasks. They're loaded on demand when the agent needs them.

How do I create a custom skill?

1. Create a directory: .localai/skills/my-skill/
2. Add SKILL.md with frontmatter and instructions
3. The skill appears automatically in the Skills panel

How do I customize the agent's behavior?

Create .localai/system.md in your workspace root with your instructions. The agent loads this on every run.

Can I use multiple LLM providers?

Yes. You can configure multiple providers and switch between them in the Agent panel.

Troubleshooting

The editor won't start

1. Ensure Node.js 22+ is installed: node --version
2. Ensure pnpm 10+ is installed: pnpm --version
3. Try a clean install: rm -rf node_modules && pnpm install

The AI isn't responding

1. Check your LLM provider is running
2. Test the connection in Settings → LLM Providers
3. Check the server logs for errors

MCP server won't connect

1. Verify the command and arguments
2. Check environment variables
3. Try running the command manually in terminal

Docker image won't build

1. Ensure Docker BuildKit is enabled: DOCKER_BUILDKIT=1
2. Check you have enough disk space
3. Try a clean build: docker build --no-cache -t localai-code-editor .

Contributing

How do I contribute?

See Contributing for guidelines.

Where do I report bugs?

Open an issue on GitHub Issues with steps to reproduce.

Where do I request features?

Start a discussion in GitHub Discussions → Ideas.