deploy: e4b277d

index.html

@@ -79,10 +79,20 @@ <h1><a class="anchor" id="autotoc_md0"></a>
 <p>This is enabled by LLM model compression technique: <a href="https://github.com/mit-han-lab/smoothquant">SmoothQuant</a> and <a href="https://github.com/mit-han-lab/llm-awq">AWQ (Activation-aware Weight Quantization)</a>, co-designed with TinyChatEngine that implements the compressed low-precision model.</p>
 <h3><a class="anchor" id="autotoc_md1"></a>
 Demo on an NVIDIA GeForce RTX 4070 laptop:</h3>
-<p><img src="assets/figures/chat_demo_gpu.gif" alt="chat_demo_gpu" width="49%" class="inline"/> <img src="assets/figures/coding_demo_gpu.gif" alt="coding_demo_gpu" width="49%" class="inline"/></p>
+<table class="doxtable">
+<tr>
+<td><img src="assets/figures/chat_demo_gpu.gif" alt="chat_demo_gpu" width="100%" class="inline"/>  </td><td><img src="assets/figures/coding_demo_gpu.gif" alt="coding_demo_gpu" width="100%" class="inline"/>   </td></tr>
+<tr>
+<td align="center">LLaMA Chat  </td><td align="center">Code LLaMA   </td></tr>
+</table>
 <h3><a class="anchor" id="autotoc_md2"></a>
 Demo on an Apple MacBook Pro (M1, 2021):</h3>
-<p><img src="assets/figures/chat_demo_m1.gif" alt="chat_demo_m1" width="49%" class="inline"/> <img src="assets/figures/coding_demo_m1.gif" alt="coding_demo_m1" width="49%" class="inline"/></p>
+<table class="doxtable">
+<tr>
+<td><img src="assets/figures/chat_demo_m1.gif" alt="chat_demo_m1" width="100%" class="inline"/>  </td><td><img src="assets/figures/coding_demo_m1.gif" alt="coding_demo_m1" width="100%" class="inline"/>   </td></tr>
+<tr>
+<td align="center">LLaMA Chat  </td><td align="center">Code LLaMA   </td></tr>
+</table>
 <p>Feel free to check out our <a href="assets/slides.pdf">slides</a> for more details!</p>
 <h2><a class="anchor" id="autotoc_md3"></a>
 Overview</h2>