Quick Start

Download the ZIP from GitHub Releases and extract it to a folder of your choice. The LFM2.5-1.2B-JP model is included — no additional download is required to get started.

Open the extracted folder and run OnecaratEditor.exe. A language selection dialog will appear on first launch.

Windows SmartScreen may display a warning on first launch. This is expected for new software. Click "More info" then "Run anyway" to proceed.

1. Open the menu Settings → LLM Settings…
2. The bundled LFM2.5-1.2B-JP model is pre-configured. To use a different model, set the model folder.
3. Select the model you want to use from the list.
4. Click the "Start Server" button.

The setup popup will close automatically once the server starts successfully.

• Chat with the AI character to brainstorm or draft ideas.
• Select text, right-click, and choose AI actions such as summarize, rewrite, or expand.

To use a different model, download a GGUF file from Hugging Face and place it in a folder of your choice (e.g. C:\work\gguf). Some models require a free Hugging Face account.

For speech-to-text in the Free edition, Whisper-related files must be prepared separately by the user.

Get OnecaratEditor Core from the Microsoft Store. Installation and updates are handled by Windows — no ZIP files, no SmartScreen warnings.

On first launch, the setup guides you through downloading an AI model from Hugging Face — three models to choose from, with the default picked from your UI language and available RAM. This is the only time the app accesses the internet — and this request, too, is recorded in network.log.

After setup, everything — chat, folder RAG, voice input, AI text processing — runs on your PC. You can verify it anytime: Help menu → "Open network log".