incorrect output

[MhL5]

Before submitting your bug report

• I believe this is a bug. I'll try to join the Continue Discord for questions
• I'm not able to find an open issue that reports the same bug
• I've seen the troubleshooting guide on the Continue Docs

Relevant environment info

- OS: windows 11
- Continue: v0.8.40
- IDE: vscode / webstorm
- LLM Model: deepseek-coder-v2:latest

Description

Hey,

This morning, I was trying to set up Continue in WebStorm. I wanted to fix something else, and I messed around with the settings and invalidated all the caches using the WebStorm IDE.

I don't know if they are related or not, but after that, I've been receiving weird responses from the model like this both in vs code and webstorm:

restarting the computer or closing and opening ollama desktop app solves this issue, but it keeps happening for some reason after working for about half an hour.

[sestinj]

@MhL5 it seems like there's likely a problem outside of Continue—are you able to make curl requests to Ollama successfully?

[MhL5]

I can use the model with ollama run deepseek-coder-v2:latest in the terminal, and it works. However, in VS Code, the output is incorrect until I close the Ollama desktop app and reopen it.

i need to investigate this more, I'm not sure where this is coming from