llama2.c is a very simple implementation to run inference of models with a Llama2-like transformer-based LLM architecture.

This is a pure C# implementation of the same thing. It is optimized for speed and very simple to understand and modify.

Requires .net7 or higher.

1. First put the stories15M.bin file in the same directory as the executable. You can download it from here
2. Get tokenizer from here and put it in the same directory as the executable.

dotnet build -c Release

.\bin\Release\net7.0\llama2.cs.exe stories15M.bin

.\bin\Release\net7.0\llama2.cs.exe stories15M.bin  -i "A long time ago a"

• Inference with Llama2 checkpoints
• Use high performance C# types from .net8?
• Add training functionality