FIM requests not being sent / recieved at Ollama API

[lewismacnow]

Describe the bug I have ollama running on a 'good' host, been used with ollama for multiple APIs for a while. With many models. Twinny extension is installed. Settings updated to point at the external ollama api. Completions via the chat window are working perfectly. FIM returns nothing. No error, no timeout. And the Logs show no requests are being recieved.

To Reproduce Steps to reproduce the behavior:

1. Install Twinny

2. Check/update settings to ensure model correct and available

3. Test co-pilot style chat

3.1. Check Logs for Ollama to see the API request:

4. Test auto code complete - Even using manual key-binding nothing happens but this loading symbol spins (forever) and never stops until I click it

4.1. Check Ollama server for API request - there are none, the request from 3.1 is still the most recent:

Expected behavior Code completions

Desktop (please complete the following information):

• MacOS 14.4 (client)
• Ubuntu 22 (Ollama Host)
• Ollama 0.1.28

Additional context Ollama autocoder extension works which has similar features to what I suspect you have created for Twinny. It seems the HTTP requets is not making it to my ollama server?

[lewismacnow]

UPDATE: This issue was caused by a setting in VS Code. (Not sure which one).

SOLUTION: I reset all VS Code settings to default;

1. Open Visual Studio Code and click on the settings icon.
2. Click on "Settings" and then on the file icon to open the settings.json file.
3. Remove all existing settings in the file, leaving only the curly braces {}.
4. Save the file. This action will reset all your settings.
5. Restart Visual Studio Code for the changes to take effect.