.NET LLaMA.cpp library! 🦙 #494
Replies: 7 comments 8 replies
-
When I run the project I get this: System.AccessViolationException: 'Attempted to read or write protected memory. This is often an indication that other memory is corrupt.' It occurs at line 26 of LLaMARunner.cs And this is the CMD window output:
|
Beta Was this translation helpful? Give feedback.
-
I have never used CMake before, so I am a total newbie. My hope is that sharing the output from the cmake command, it would let you spot my error. Thank you sir! I am not sure that this matters, but I am using Visual Studio 2022 Community Edition, and my dev system is an AMD processor.
I am not sure if the hardware matters for this project? |
Beta Was this translation helpful? Give feedback.
-
I opened that question for C++ help as I think the problem is with the llama.cpp build. |
Beta Was this translation helpful? Give feedback.
-
Nice work! Looking forward to future updates, thanks for sharing. I have been playing around with using llama.cpp in dotnet on my end as well and I created a quick project for fun - if you want to check it out. I am essentially compiling main.exe as a main.dll instead and then using DllImport then I added two hook functions (one for fetching the next input and another for the next output). |
Beta Was this translation helpful? Give feedback.
-
Hi all, I hope I could have seen this discussion earlier. We also made a repo LLamaSharp and have released NuGet packages to allow users to use it without compiling themselves. It provides many features and will be long-time maintained by SciSharp community. May I kindly ask that if you would like to develop it together with us since you all are also experienced with LLM in dotnet? Currently there are many feature requests so the amount of the work is a little big for me. I think gathering the community effort will make the LLama toolchain in dotnet better. I'll be very glad and many thanks if you'd like to make it together 😊 |
Beta Was this translation helpful? Give feedback.
-
Very cool! I was going to use it today but it seems to be 3 months behind llama.cpp. Care to update it? |
Beta Was this translation helpful? Give feedback.
-
Got bored today and slapped together a .NET Core library wrapped around llama.cpp! Got inference working, but still need to work through the model conversions 😁
Beta Was this translation helpful? Give feedback.
All reactions