Update README

README.md

@@ -46,6 +46,13 @@ library_name: transformers
   <a href="https://github.com/deepseek-ai/DeepSeek-R1/blob/main/DeepSeek_R1.pdf"><b>Paper Link</b>👁️</a>
 </p>
 
+# GGUF and Quantized versions of deepseek-ai/DeepSeek-R1-Distill-Qwen-14B
+This is a fork of [deepseek-ai/DeepSeek-R1-Distill-Qwen-14B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-14B) where the safetensors have been converted to GGUF and quantized to BF16, Q8_0, and Q4_K
+
+This model seems to perform really well in reasoning and text generation tasks. Given how the [DeepSeek](https://www.deepseek.com/) team managed to create and train the R1 models in a remarkable cost efficient way, it is a major achievement in the field!
+
+From the original repo:
+
 ## 1. Introduction
 
 We introduce our first-generation reasoning models, DeepSeek-R1-Zero and DeepSeek-R1.