Update README.md

README.md

@@ -121,7 +121,12 @@ code {
          The model stock merging method attempts to make the model remain focused, tailored, and high-quality.
       <h2>Quants:</h2>
     <p>(Thanks to <a href="https://huggingface.co/mradermacher">@Mradermacher!</a>, please send them likes and follows!)</p>
-    <p><a href="https://huggingface.co/mradermacher/L3-MS-Astoria-70b-GGUF">L3-MS-Astoria-70b-GGUF (GGUFs)</a></p>        
+    <p><a href="https://huggingface.co/mradermacher/L3-MS-Astoria-70b-GGUF">L3-MS-Astoria-70b-GGUF (GGUFs)</a></p>    
+    <p><h2> ExLlamaV2 Quantization by UnstableLlama </h2></p>
+
+      [5.00 bits per weight in main](https://huggingface.co/UnstableLlama/Xwin-LM-13B-V0.2-exl2/tree/main)
+
+      [3.00 bits per weight branch](https://huggingface.co/UnstableLlama/Xwin-LM-13B-V0.2-exl2/tree/3bpw)    
     <p></p>  
       <h3>Config:</h3>
       <pre><code>MODEL_NAME = "L3-MS-Astoria-70b"