How to replace LlamaSharp nuget with git project?

[n-ate]

Hi! I've been running a project against Llamasharp for a couple of months now. I'm using the nuget package with the CUDA12 Windows backend package. I want to dig more into the Llamasharp code to understand it beter and the best way for me to do this is to swap out the nuget package with the libraries from github. The issue with this approach is that I get dll binding issues against llama.dll. I imagine that I need to specify which lama.dll to use somewhere and where it is. Can someone please point me in the right direction to get the project running on my local? Thanks!

[martindevans]

The main project should auto download everything necessary, see here in the csproj.

I think you probably just need to add a line like this one in the example project to tell it where to get those binaries from.