add Phi-2 to README.md

README.md

@@ -31,7 +31,7 @@ LLMUnity is built on top of the awesome [llama.cpp](https://github.com/ggerganov
 
 - 💬 Craft your own NPC AI that the player interacts with in natural language
 - 🖌️ The NPC can respond creatively according to your character definition
-- 🗝️ ... or with predefined answers that stick to your script!
+- 🗝️ or with predefined answers that stick to your script!
 - 🔍 You can even build an intelligent search engine with our search system (RAG)
 
 **Features**
@@ -347,7 +347,7 @@ In the scene, select the `LLM` GameObject download the default model (`Download
 Save the scene, run and enjoy!
 
 ## Use your own model
-LLMUnity uses the Mistral 7B Instruct model by default, quantised with the Q4 method ([link](https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.2-GGUF/resolve/main/mistral-7b-instruct-v0.2.Q4_K_M.gguf?download=true)).<br>
+LLMUnity uses the [Microsoft Phi-2](https://huggingface.co/microsoft/phi-2) or [Mistral 7B Instruct](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) model by default, quantised with the Q4 method.<br>
 Alternative models can be downloaded from [HuggingFace](https://huggingface.co/models).<br>
 The required model format is .gguf as defined by the llama.cpp.<br>
 The easiest way is to download gguf models directly by [TheBloke](https://huggingface.co/TheBloke) who has converted an astonishing number of models :rainbow:!<br>