Update README.md

README.md

@@ -388,21 +388,21 @@ The following performance benchmarks were conducted with [vLLM](https://docs.vll
       <td>neuralmagic/Qwen2-VL-72B-Instruct-quantized.w8a8</td>
       <td>1.84</td>
       <td>1.2</td>
-      <td>293</td>
+      <td>586</td>
       <td>4.0</td>
-      <td>1021</td>
+      <td>2042</td>
       <td>4.6</td>
-      <td>1135</td>
+      <td>2270</td>
     </tr>
     <tr>
       <td>neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16</td>
       <td>2.73</td>
       <td>2.4</td>
-      <td>314</td>
+      <td>1256</td>
       <td>12.8</td>
-      <td>1591</td>
+      <td>6364</td>
       <td>16.0</td>
-      <td>2019</td>
+      <td>8076</td>
     </tr>
     <tr>
       <th rowspan="3" valign="top">H100x4</td>
@@ -419,21 +419,21 @@ The following performance benchmarks were conducted with [vLLM](https://docs.vll
       <td>neuralmagic/Qwen2-VL-72B-Instruct-FP8-Dynamic</td>
       <td>1.70</td>
       <td>1.6</td>
-      <td>236</td>
+      <td>457</td>
       <td>4.4</td>
-      <td>623</td>
+      <td>1207</td>
       <td>4.8</td>
-      <td>669</td>
+      <td>1296</td>
     </tr>
     <tr>
       <td>neuralmagic/Qwen2-VL-72B-Instruct-quantized.w4a16</td>
       <td>2.35</td>
       <td>5.2</td>
-      <td>350</td>
+      <td>1400</td>
       <td>13.2</td>
-      <td>910</td>
+      <td>3640</td>
       <td>14.4</td>
-      <td>994</td>
+      <td>3976</td>
     </tr>
   </tbody>
 </table>