There is a problem using ollama in vscode+jupyter #184

Closed
opened 2026-02-15 16:28:33 -05:00 by yindo · 1 comment
Owner

Originally created by @szzhh on GitHub (Dec 12, 2024).

I am using Windows system locally. I use vscode+jupyter to remotely connect to Linux server and start ollama service. When I run ollama.list(), the model output is the model downloaded locally. However, when I run the magic command! ollama list command, it outputs the model of the remote server. And the jupyter I created remotely can only use the local ollama model, not the remote model.
PixPin_2024-12-12_19-29-47

Originally created by @szzhh on GitHub (Dec 12, 2024). I am using Windows system locally. I use vscode+jupyter to remotely connect to Linux server and start ollama service. When I run ollama.list(), the model output is the model downloaded locally. However, when I run the magic command! ollama list command, it outputs the model of the remote server. And the jupyter I created remotely can only use the local ollama model, not the remote model. ![PixPin_2024-12-12_19-29-47](https://github.com/user-attachments/assets/da3933e2-526a-4f51-bcdb-e884fcefe39b)
yindo closed this issue 2026-02-15 16:28:33 -05:00
Author
Owner

@ParthSareen commented on GitHub (Dec 12, 2024):

Hi @szzhh - to do this you can either to pass in your own http client or export OLLAMA_HOST=HOSTNAME to point Ollama to the right spot

@ParthSareen commented on GitHub (Dec 12, 2024): Hi @szzhh - to do this you can either to pass in your own http client or `export OLLAMA_HOST=HOSTNAME` to point Ollama to the right spot
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: ollama/ollama-python#184