Help: Ollama Setup

Migrated from lrgenius.com/help/ollama-setup and curated for repo docs.
Screenshot references were intentionally removed.

1. Install Ollama

• Download from: https://ollama.com/
• Install for your platform (Windows/macOS/Linux as available)

2. Pull at least one vision-capable model

For current model recommendations and hardware sizing guidance, see Help: Choosing AI Model.

You can browse all available vision models here:

• https://ollama.com/search?c=vision — each model page shows the required VRAM per quantisation variant.

3. Nothing to configure in the plugin

LrGeniusAI finds Ollama on this computer by itself, at its default address (http://localhost:11434). Once Ollama is running, its models appear in every task’s AI Model list as Ollama · <model>.

Ollama on another computer (or on a changed port): enter its address, e.g. http://nas.local:11434, as the Other AI server — see Other AI Server. That connection goes through Ollama’s OpenAI-compatible API, which cannot set the context size per request; raise it on that computer if long prompts get cut off (OLLAMA_CONTEXT_LENGTH=8192 ollama serve).

Notes

• Larger models generally improve quality but need more VRAM/RAM.
• First pull can take significant time due to model size.