The notice says semantic search is paused

Semantic search uses the nomic-embed-text model through Ollama; when it is missing, or Ollama is not running, indexing pauses and the notice says which, while chat and keyword search keep working.

Last checked against the app on · View as Markdown

Steps

  1. Read the notice. It has three forms. "The nomic-embed-text model is not installed" means Ollama is up but the model was never pulled. "Ollama is not running" means nothing answered. "Ollama rejected a request for the nomic-embed-text model" means it is up and still refusing, with the detail at the end.
  2. For a missing model, run "ollama pull nomic-embed-text" in a terminal.
  3. For a stopped Ollama, start it.
  4. Indexing resumes on its own once the model is present and Ollama answers. There is nothing to press in NeuralVault.

How to tell it worked

The notice goes away and semantic search returns results after indexing catches up.

If it does not work

You started Ollama and the notice is still there.
Give indexing time to resume. If it persists, confirm the model is present by running "ollama list".
You pulled the model but Ollama still rejects the request.
The notice ends with the detail Ollama returned. Quote it in a bug report. The notice cannot name the cause beyond that.
Chat or keyword search seems broken too.
They are not affected by this notice. If they fail, see the guide on a model not answering.