✳

Connect your local Ollama

Your chat runs on your machine. Retrieval on ours.

Checking... ✓ models available

Ollama not reachable. Set it up:

  1. Install Ollama: ollama.com/download
  2. Pull a chat model:
    ollama pull qwen2.5:7b-instruct
  3. Allow this domain in Ollama CORS (macOS):
    launchctl setenv OLLAMA_ORIGINS "*"

    then restart Ollama.

  4. Linux/Windows: OLLAMA_ORIGINS=* ollama serve

Datasets

Upload your documents. Retrieval indexes them for chat.

Create new
A
bring-your-own-Ollama

Retrieval on ask.oliverkoehn.com · LLM on · Chat never persists on server.