From ee89b24879fb65ca9e1c6d043e641c23de6bbfce Mon Sep 17 00:00:00 2001
From: latent-9 <296084221+latent-9@users.noreply.github.com>
Date: Sat, 1 Aug 2026 05:51:47 +1200
Subject: [PATCH] docs: fix example config paths in English installation guide
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The English quick-start referenced examples/v1/sft_qwen3_tiny.py and
examples/v1/sft_intern_s1_tiny_config.py, which 404 — the configs live under
examples/v1/config/. The Chinese installation guide already uses the correct
examples/v1/config/ paths; this brings the English doc in line so the
documented torchrun commands run.
---
 docs/en/get_started/installation.md | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/docs/en/get_started/installation.md b/docs/en/get_started/installation.md
index 2e8f569d1f..20feefac98 100644
--- a/docs/en/get_started/installation.md
+++ b/docs/en/get_started/installation.md
@@ -68,7 +68,7 @@ Having problems running? Check out the [FAQ](faq)
 :caption: dense model fine-tuning example
 :linenos:
 
-torchrun  xtuner/v1/train/cli/sft.py --model-cfg examples/v1/sft_qwen3_tiny.py --chat_template qwen3 --dataset tests/resource/openai_sft.jsonl
+torchrun  xtuner/v1/train/cli/sft.py --model-cfg examples/v1/config/sft_qwen3_tiny.py --chat_template qwen3 --dataset tests/resource/openai_sft.jsonl
 ```
 
 After successful execution, the log is as follows
@@ -91,7 +91,7 @@ After successful execution, the log is as follows
 ```
 ````
 
-The above log shows that only 8G of memory is needed to run. If you want to reduce memory usage further, you can consider modifying the `num_hidden_layers` and `hidden_size` parameters in `examples/v1/sft_qwen3_tiny.py`.
+The above log shows that only 8G of memory is needed to run. If you want to reduce memory usage further, you can consider modifying the `num_hidden_layers` and `hidden_size` parameters in `examples/v1/config/sft_qwen3_tiny.py`.
 
 ### MLLM Multimodal Large Model Fine-tuning
 
@@ -103,7 +103,7 @@ Take Intern-S1 scientific multimodal as an example
 :caption: Intern-S1 tiny model fine-tuning example
 :linenos:
 
-torchrun xtuner/v1/train/cli/sft.py --config examples/v1/sft_intern_s1_tiny_config.py
+torchrun xtuner/v1/train/cli/sft.py --config examples/v1/config/sft_intern_s1_tiny_config.py
 ```
 
 After successful execution, the log is as follows
@@ -267,7 +267,7 @@ WARNING: input_ids length 4171 exceeds model_max_length 4096. truncated!
 ```
 ````
 
-The above log shows that only 10G of memory is needed to run. If you want to reduce memory usage further, you can consider modifying the `llm_cfg` dictionary related parameters in `examples/v1/sft_intern_s1_tiny_config.py`.
+The above log shows that only 10G of memory is needed to run. If you want to reduce memory usage further, you can consider modifying the `llm_cfg` dictionary related parameters in `examples/v1/config/sft_intern_s1_tiny_config.py`.
 
 (faq)=
 ## FAQ