Quick Start

Get up and running with Karpa in just a few minutes. Pick one of the supported providers below.

Prerequisites

Pick one of the following providers:

• LM Studio — Download, load a model (HY-MT1.5-7B recommended), and start the local server on port 1234.
• Ollama — Download and pull a model (e.g. ollama pull llama3.2).
• OpenAI / Anthropic / Google Gemini / OpenRouter — Create an account and get an API key.

Run Karpa

Docker (recommended)

git clone https://github.com/sudoeren/karpa.git && cd karpa
docker compose up -d

Development

git clone https://github.com/sudoeren/karpa.git && cd karpa
npm install
npm run dev

Open http://localhost:7250 and start translating.

Quick sanity check

1. Start your provider (LM Studio server, Ollama, or configure API key).
2. Open Karpa's Settings → Connection to select your provider.
3. Click Test Connection to verify.
4. Try a translation.

If Karpa can't reach your provider, check the environment variables or in-app settings.

Next steps

• Check all installation options (Docker, dev, Docker pull, custom build).
• Configure your provider settings.
• Learn text translation and file translation.