Skip to content

Ollama Setup

Ollama connects Ebla AI to models hosted on your own infrastructure.

Prepare Ollama

  1. Install Ollama on a server reachable from EspoCRM.
  2. Pull the required model, for example:
ollama pull llama3
  1. Confirm the Ollama API is reachable. Its common local URL is http://localhost:11434.

Connect EspoCRM

  1. Open Administration > Integrations > AI » Ollama.
  2. Enable the integration.
  3. Enter the API base URL.
  4. Click Test & Fetch Models.
  5. Select a model and test again to verify generation.
  6. Save.
  7. Select Ollama under Administration > AI Settings > General when it should be the default provider.

Ollama integration

The Ollama server is authoritative for available local models. Advanced capabilities depend on the selected model and the transport supported by the extension.

Production notes

  • Do not expose an unauthenticated Ollama endpoint to the public internet.
  • Restrict network access between EspoCRM and the Ollama server.
  • Monitor model memory, response time, and context limits.
  • Test CRM tool use and file inputs before granting them to users.