Update README.md

README.md

@@ -28,7 +28,7 @@ tags:
 - **Model Developers:** Neural Magic
 
 Quantized version of [Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct).
-It achieves an average score of 43.38 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark (version 1), whereas the unquantized model achieves 43.64.
+It achieves an average score of 43.38 on the [OpenLLM](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard) benchmark version 1 and 23.42 on version 2, whereas the unquantized model achieves 43.64 on version 1 and 23.39 on version 2.
 
 ### Model Optimizations
 
@@ -95,6 +95,8 @@ lm_eval \
    </td>
   </tr>
   <tr>
+   <td rowspan="7" ><strong>OpenLLM v1</strong>
+   </td>
    <td>MMLU (5-shot)
    </td>
    <td>46.83
@@ -164,5 +166,77 @@ lm_eval \
    <td><strong>99.4%</strong>
    </td>
   </tr>
+  <tr>
+   <td rowspan="7" ><strong>OpenLLM v2</strong>
+   </td>
+   <td>MMLU-Pro (5-shot)
+   </td>
+   <td>17.49
+   </td>
+   <td>16.95
+   </td>
+   <td>96.9%
+   </td>
+  </tr>
+  <tr>
+   <td>IFEval (0-shot)
+   </td>
+   <td>31.17
+   </td>
+   <td>32.04
+   </td>
+   <td>102.8%
+   </td>
+  </tr>
+  <tr>
+   <td>BBH (3-shot)
+   </td>
+   <td>32.79
+   </td>
+   <td>32.51
+   </td>
+   <td>99.2%
+   </td>
+  </tr>
+  <tr>
+   <td>Math-lvl-5 (4-shot)
+   </td>
+   <td>0.21
+   </td>
+   <td>0.17
+   </td>
+   <td>***
+   </td>
+  </tr>
+  <tr>
+   <td>GPQA (0-shot)
+   </td>
+   <td>25.67
+   </td>
+   <td>26.12
+   </td>
+   <td>101.8%
+   </td>
+  </tr>
+  <tr>
+   <td>MuSR (0-shot)
+   </td>
+   <td>33.02
+   </td>
+   <td>32.75
+   </td>
+   <td>99.2%
+   </td>
+  </tr>
+  <tr>
+   <td><strong>Average</strong>
+   </td>
+   <td><strong>23.39</strong>
+   </td>
+   <td><strong>23.42</strong>
+   </td>
+   <td><strong>100.1%</strong>
+   </td>
+  </tr>
 </table>
-
+*** Reference value too low to report meaningful recovery.