From 46994b34fb5231b00328f19ed9c6f784b9d61db1 Mon Sep 17 00:00:00 2001
From: Baptiste Colle <32412211+baptistecolle@users.noreply.github.com>
Date: Wed, 15 Jan 2025 16:26:11 +0100
Subject: [PATCH] :memo: add guide on using TPU with TGI in the docs (#2907)

---
 docs/source/_toctree.yml        | 2 ++
 docs/source/installation_tpu.md | 3 +++
 2 files changed, 5 insertions(+)
 create mode 100644 docs/source/installation_tpu.md

diff --git a/docs/source/_toctree.yml b/docs/source/_toctree.yml
index 4503424bb..ab85682b7 100644
--- a/docs/source/_toctree.yml
+++ b/docs/source/_toctree.yml
@@ -13,6 +13,8 @@
     title: Using TGI with Intel Gaudi
   - local: installation_inferentia
     title: Using TGI with AWS Inferentia
+  - local: installation_tpu
+    title: Using TGI with Google TPU
   - local: installation_intel
     title: Using TGI with Intel GPUs
   - local: installation
diff --git a/docs/source/installation_tpu.md b/docs/source/installation_tpu.md
new file mode 100644
index 000000000..208ebce3c
--- /dev/null
+++ b/docs/source/installation_tpu.md
@@ -0,0 +1,3 @@
+# Using TGI with Google TPU
+
+Check out this [guide](https://huggingface.co/docs/optimum-tpu) on how to serve models with TGI on TPUs.