From 491e295a48ccddbfe98b74e39e4e350aa8c5acd7 Mon Sep 17 00:00:00 2001
From: Meng Zhang <meng@tabbyml.com>
Date: Sun, 10 Sep 2023 22:16:46 +0800
Subject: [PATCH] docs: add faq on utilize multipl NV GPUs (#423)

* docs: add faq on utilize multipl NV GPUs

* update
---
 website/docs/{03-faq.md => 03-faq.mdx} | 13 +++++++++++++
 1 file changed, 13 insertions(+)
 rename website/docs/{03-faq.md => 03-faq.mdx} (60%)

diff --git a/website/docs/03-faq.md b/website/docs/03-faq.mdx
similarity index 60%
rename from website/docs/03-faq.md
rename to website/docs/03-faq.mdx
index ed2d2ff..682abb3 100644
--- a/website/docs/03-faq.md
+++ b/website/docs/03-faq.mdx
@@ -1,3 +1,5 @@
+import CodeBlock from '@theme/CodeBlock';
+
 # ⁉️ Frequently Asked Questions
 
 <details>
@@ -18,3 +20,14 @@
     </p>
   </div>
 </details>
+
+<details>
+  <summary>How to utilize multiple NVIDIA GPUs?</summary>
+  <div>
+    <p>Tabby supports replicating models on multiple GPUs to increase throughput. You can specify the devices for model replication by using the <b>--device-indices</b> option.</p>
+    <CodeBlock language="bash">
+    # Replicate model to GPU 0 and GPU 1.{'\n'}
+    tabby serve ... --device-indices 0 --device-indices 1
+    </CodeBlock>
+  </div>
+</details>