Doesn't work with local-hosted Ollama on Linux

[5a9awneh]

Describe the bug Interface slow to load, when it does, I receive Invalid connection message, in settings (also slow to load) it doesn't detect installed Ollama models

To Reproduce Steps to reproduce the behavior:

1. Install Ollama locally (not as docker)
2. Tried different Ollama APIs in config.toml: http://host.docker.internal:11434 http://localhost:11434 http://private_ip_of_computer_hosting_ollama:11434
3. Install Perplexica via docker compose up -d
4. Browse to http://localhost:3000 to See error
5. Open Settings, doesn't detect neither Ollama nor Ollama models

Expected behavior It should be able to connect to Ollama and detect installed models

Additional context

• System: Linux Mint
• Ollama (Local) is already working fine in both Terminal & Open WebUI (docker)

[ItzCrazyKns]

You need to expose Ollama to your network and then use the private IP of the computer + port as the API URL. Feel free to re-open this issue or join our Discord server if you face any issues after performing the steps