Skip to main content

AI commands

Overview​

Writeopia integrates AI to enhance the writing experience by providing intelligent assistance while maintaining user privacy. Users can select from different AI agents to assist with brainstorming, editing, and content generation.

Features​

  • Offline AI Support: AI functionalities work offline to ensure privacy and security.
  • Real-time Collaboration: AI can assist in real-time without interfering with user input.
  • Secure Processing: Data remains private and is not sent to external servers.

Using AI in Writeopia​

Writeopia supports any local LLM runner that exposes an Ollama-compatible API, so you're not locked into a single tool as long as it speaks the same interface.

After installing the tools like Ollama, llmman, it should work out of the box, but you can follow next sessions to guides of installation and integrations.

Ollama​

Installing Ollama​

In order to provide local AI features, Writeopia uses Ollama. Please go to Ollama website and:

  • Install Ollama at ollama.com
  • Start Ollama in your computer.

Starting Ollama is necessary for Writeopia communicate with it for the AI features.

Configuring Writeopia for Ollama.​

Writeopia can use many different LLM models. You can choose which one you prefer. Bigger models will provide better results, but they are also heavier to use for your computer. You should experiment with the models and decides which one is the best for you. You can search for many at: https://ollama.com/search.

If you are unsure about the model, no problem. We recommend that you start with gemma3:4b.

You can follow this video to install it:

llmman​

llmman is a local model runner that serves the Ollama API (alongside OpenAI- and Anthropic-compatible ones) on port 17434.

Instaling llmman​

  • Install it: curl -fsSL https://raw.githubusercontent.com/llmmanorg/llmman/main/install.sh | sh (Linux/macOS) or irm https://raw.githubusercontent.com/llmmanorg/llmman/main/install.ps1 | iex (Windows).
  • Start the server with llmman serve.
  • You can choose your models inside Writeopia interface, but you would like to start already with a model downloaded pull a model with llmman pull gemma4.
  • In Writeopia, open Settings and change the AI URL from to http://localhost:17434.

Models can be pulled from any OCI registry or straight from Hugging Face (for example hf.co/unsloth/Qwen3.5-0.8B-GGUF), and downloading, listing and deleting models from Writeopia's settings work the same way as the CLI. Models downloaded using the CLI will be available in Writeopia.

Selecting an AI Agent​

If you download more models. You will be able to select different models are use them as you prefer.

Ask AI​

To ask AI you can move the cursor to the line that you would like to make a question of select all the lines that you would like to include in the query.


There are different way to ask AI to generate text. You can find more about it in the AI section.

Future Enhancements​

We plan to add:

  • Automatic configuration for Ollama and llmman. Just click and it is configured.
  • Presentations generated by AI
  • Semantic search
  • And more!