add shields

Author: getnamo

README.md

@@ -1,5 +1,8 @@
 # Llama Unreal
 
+[![GitHub release](https://img.shields.io/github/release/getnamo/Llama-Unreal.svg)](https://github.com/getnamo/Llama-Unreal/releases)
+[![Github All Releases](https://img.shields.io/github/downloads/getnamo/Llama-Unreal/total.svg)](https://github.com/getnamo/Llama-Unreal/releases)
+
 An Unreal plugin for [llama.cpp](https://github.com/ggml-org/llama.cpp) to support embedding local LLMs in your projects.
 
 Fork is modern re-write from [upstream](https://github.com/mika314/UELlama) to support latest API, including: GPULayers, advanced sampling (MinP, Miro, etc), Jinja templates, chat history, partial rollback & context reset, regeneration, and more. Defaults to Vulkan build on windows for wider hardware support at about ~10% perf loss compared to CUDA backend on token generation speed.