Successfully running Ollama

I'd like to note a win for me. I have Ollama setup and running on an older server with an 8GB VRAM RTX 2070. In the first bit of going through foundation work I have successfully setup VS Code to look at the Ollama models and use the email skill that I created.

Next step is to see if I can get my local machine and the server working at the same time. This way I can select between differnt models and compute power. My local machine has an RX 9060 XT with 16GB VRAM. This would at least let me load a larger model locally.

I still have a great deal of building out the ICM, but I feel I'm well on my way to pratical applicaiton now. I also am not sure if I want to bother with the built in agent settings for VS Code as it doesn't seem as portable. I'd welcome anyone's thoughts on building out the local user agent settings vs using a git repo and importing it in to a workspace.

4 comments

Successfully running Ollama