Docs
Docs Troubleshooting Local AI Connections

Troubleshooting Local AI Connections

Last updated: August 10, 2026

If you’ve followed the setup guides for Ollama or LM Studio but Eviory OS still throws a “Network Error” or “Failed to fetch”, the issue is likely due to your browser’s strict security policies regarding Mixed Content.

Even with CORS properly configured, browsers (like Chrome, Edge, or Brave) often block the request because a secure site (https://eviory.com) is trying to contact an insecure endpoint (http://localhost:11434).

Here is how to fix it:

Fix 1: Use 127.0.0.1 instead of localhost

Sometimes browsers treat the IP address 127.0.0.1 more leniently than the word localhost.

In Eviory OS Settings, make sure your Local Base URL is set to: http://127.0.0.1:11434/v1 (for Ollama) rather than http://localhost:11434/v1.

Fix 2: Allow Insecure Localhost (Chrome & Edge)

If it’s still blocked, you can explicitly tell your Chromium-based browser to allow it:

  1. Open a new tab in your browser.
  2. Go to chrome://flags/#allow-insecure-localhost (or edge://flags/#allow-insecure-localhost if using Edge).
  3. Set Allow invalid certificates for resources loaded from localhost to Enabled.
  4. Restart your browser completely.

Fix 3: Disable Brave Shields

If you are using the Brave browser, the strict “Shields” feature will aggressively block localhost requests from external sites.

  1. Click the Brave Shields icon (the lion head) in the URL bar while on eviory.com.
  2. Disable the shields for this site entirely, or explicitly allow mixed content in the shield settings.

Fix 4: Verify CORS Configuration

If you’ve done the above and it still fails, the Local AI server itself might be rejecting the request.

Double check that you have correctly set OLLAMA_ORIGINS="https://eviory.com" (for Ollama) or checked the CORS box (for LM Studio) and restarted the application.