Ollama, LM Studio and custom servers

PDF Pals can answer with Ollama, LM Studio or any OpenAI-compatible server. v2 also has built-in local models.

Ollama

  1. Install Ollama, download a model and keep Ollama running.
  2. In PDF Pals, open Settings ▸ AI Providers, click +, choose Ollama, then Next.
  3. Leave Base URL empty for http://localhost:11434/v1. No API key is needed.
  4. Click Next, choose models and click Add Provider.

LM Studio and other OpenAI-compatible servers

  1. In LM Studio, load a model and start its local server.
  2. In PDF Pals, open Settings ▸ AI Providers, click +, choose OpenAI-Compatible, then Next.
  3. Enter the Base URL, including any version path. For LM Studio that's usually http://localhost:1234/v1.
  4. Add an API key if the server needs one.
  5. Click Next. If no models are listed, click Add Models Manually… (one model ID per line). If the connection test fails but the server works, click Skip Test.
  6. Click Add Provider.

Ollama, and servers on localhost, 127.0.0.1 or ::1, keep working in Local mode. Other servers count as cloud providers.