Update README.md

README.md

@@ -20,6 +20,6 @@ quantized_by: Hugi-R
 
 **Model creator**: [Multimodal Art Projection Research Community](https://huggingface.co/m-a-p)
 
-WIP, the list of quants will grow
+PAUSED. DeepSeek model are not well supported by llama.cpp https://github.com/ggerganov/llama.cpp/issues/5981
 
 Quants are tested to run with llama.cpp b2251. Quants are not yet evaluated, actual quality is unknown.