Update README.md

README.md

@@ -1,10 +1,10 @@
 ---
 license: mit
-library_name: transformers
+library_name: exllamav2
 language:
   - en
 base_model:
-  - deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+  - Zyphra/ZR1-1.5B
 datasets:
   - AI-MO/NuminaMath-CoT
   - codeparrot/apps
@@ -13,6 +13,23 @@ datasets:
   - MatrixStudio/Codeforces-Python-Submissions
 pipeline_tag: text-generation
 ---
+# ZR1-1.5B-exl2
+Original model: [ZR1-1.5B](https://huggingface.co/Zyphra/ZR1-1.5B) by [Zyphra](https://huggingface.co/Zyphra)  
+Based on: [DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B) by [DeepSeek](https://huggingface.co/deepseek-ai)  
+Foundation model: [Qwen2.5-1.5B](https://huggingface.co/Qwen/Qwen2.5-1.5B) by [Qwen](https://huggingface.co/Qwen)
+
+## Quants
+[4bpw h6 (main)](https://huggingface.co/cgus/ZR1-1.5B-exl2/tree/main)  
+[4.5bpw h6](https://huggingface.co/cgus/ZR1-1.5B-exl2/tree/4.5bpw-h6)  
+[5bpw h6](https://huggingface.co/cgus/ZR1-1.5B-exl2/tree/5bpw-h6)  
+[6bpw h6](https://huggingface.co/cgus/ZR1-1.5B-exl2/tree/6bpw-h6)  
+[8bpw h8](https://huggingface.co/cgus/ZR1-1.5B-exl2/tree/8bpw-h8)  
+
+## Quantization notes
+Made with Exllamav2 0.2.8 with default dataset.  
+This model can be used with TabbyAPI or Text-Generation-WebUI with RTX GPU on Windows or RTX/ROCm on Linux.  
+
+# Original model card
 # ZR1-1.5B
 
 ZR1-1.5B is a small reasoning model trained extensively on both verified coding and mathematics problems with reinforcement learning. The model outperforms Llama-3.1-70B-Instruct on hard coding tasks and improves upon the base R1-Distill-1.5B model by over 50%, while achieving strong scores on math evaluations and a 37.91% pass@1 accuracy on GPQA-Diamond with just 1.5B parameters.