From ae32df1f834a50e34ddd0fae9ea9ea2f4b092eb4 Mon Sep 17 00:00:00 2001
From: Mergen Nachin <mnachin@meta.com>
Date: Sat, 9 Aug 2025 10:08:21 +0100
Subject: [PATCH] Add documentation for pytorch_tokenizer missing

---
 docs/source/llm/export-llm.md        | 10 ++++++++++
 docs/source/using-executorch-faqs.md |  7 +++++++
 2 files changed, 17 insertions(+)

diff --git a/docs/source/llm/export-llm.md b/docs/source/llm/export-llm.md
index 35f17a8aa72..462d9a51849 100644
--- a/docs/source/llm/export-llm.md
+++ b/docs/source/llm/export-llm.md
@@ -2,6 +2,16 @@
 
 Instead of needing to manually write code to call torch.export(), use ExecuTorch's assortment of lowering APIs, or even interact with TorchAO quantize_ APIs for quantization, we have provided an out of box experience which performantly exports a selection of supported models to ExecuTorch.
 
+## Prerequisites
+
+The LLM export functionality requires the `pytorch_tokenizers` package. If you encounter a `ModuleNotFoundError: No module named 'pytorch_tokenizers'` error, install it from the ExecutorTorch source code:
+
+```bash
+pip install -e ./extension/llm/tokenizers/
+```
+
+## Supported Models
+
 As of this doc, the list of supported LLMs include the following:
 - Llama 2/3/3.1/3.2
 - Qwen 2.5/3
diff --git a/docs/source/using-executorch-faqs.md b/docs/source/using-executorch-faqs.md
index f639524d69c..d1bd0390569 100644
--- a/docs/source/using-executorch-faqs.md
+++ b/docs/source/using-executorch-faqs.md
@@ -14,6 +14,13 @@ sudo apt install python<version>-dev
 ```
 if you are using Ubuntu, or use an equivalent install command.
 
+### ModuleNotFoundError: No module named 'pytorch_tokenizers'
+
+The `pytorch_tokenizers` package is required for LLM export functionality. Install it from the ExecutorTorch source code:
+```
+pip install -e ./extension/llm/tokenizers/
+```
+
 ## Export
 
 ### Missing out variants: { _ }