Skip to content
M.A.V.E.Ndocs

Local

Local is the offline Chat engine for any local Ollama model, in Chat - /chat?engine=local.

Local - M.A.V.E.N

What it does

Section titled “What it does”

Chats against any model you have pulled with Ollama, with a model dropdown and a build mode that generates HTML, plus preview and saved builds. It runs fully offline and uses the shared Chat / History / Workspace tabs and Settings panel. Anything you pull locally shows up here - the cloud Qwen and DeepSeek engines are separate.

Key actions

Section titled “Key actions”
  • Model dropdown - pick any installed Ollama model.
  • Build mode - generate a single-file HTML artifact.
  • Preview / saved builds - browse and preview what you built.

Setup

Section titled “Setup”

Requires Ollama plus at least one pulled model. Check with command -v ollama; install from ollama.com/download. Pull a general model, for example:

Terminal window
ollama pull gemma2

No API key is needed - everything stays on the machine.

Tips

Section titled “Tips”
  • Ollama also backs Loop’s local judge and the local Qwen/DeepSeek models in the AI Mastermind.

See also

Section titled “See also”