diff --git a/.typos.toml b/.typos.toml
index 2b5850bbc..d73a98895 100644
--- a/.typos.toml
+++ b/.typos.toml
@@ -8,7 +8,8 @@ extend-ignore-identifiers-re = [
     "noe",
     "NOE",
     "BA",
-    "lin2-rin2"
+    "lin2-rin2",
+    "SME"
 ]
 extend-ignore-re = [
     "lin2-rin2",
diff --git a/docs/common/ai/_llama_cpp.mdx b/docs/common/ai/_llama_cpp.mdx
index 8ab8197aa..a46f40122 100644
--- a/docs/common/ai/_llama_cpp.mdx
+++ b/docs/common/ai/_llama_cpp.mdx
@@ -11,7 +11,7 @@ llama.cpp 是一个基于纯 C/C++ 实现的高性能大模型推理框架，它
 <NewCodeBlock tip="Device" type="device">
 
 ```bash
-git clone https://github.com/ggml-org/llama.cpp.git
+git clone https://github.com/ggml-org/llama.cpp.git && cd llama.cpp
 ```
 
 </NewCodeBlock>
@@ -23,7 +23,7 @@ git clone https://github.com/ggml-org/llama.cpp.git
 <NewCodeBlock tip="Device" type="device">
 
 ```bash
-sudo apt install cmake gcc g++
+sudo apt install cmake gcc g++ libcurl4-openssl-dev
 ```
 
 </NewCodeBlock>
@@ -39,15 +39,18 @@ cmake --build build --config Release -j$(nproc)
 
 </NewCodeBlock>
 
-:::info[KleidiAI]
+:::info[ARMv9]
 对于采用 ARM-v9 架构的 [瑞莎星睿 O6 / O6N](/orion/o6) 设备，可以开启 armv9-a 和 KleidiAI 编译选项进行硬件级优化。
+
+**请使用 `4aced7a` commit**
 :::
 
 <NewCodeBlock tip="Device" type="device">
 
 ```bash
+git checkout 4aced7a
 cmake -B build -DGGML_NATIVE=OFF -DGGML_CPU_ARM_ARCH=armv9-a+i8mm+dotprod -DGGML_CPU_KLEIDIAI=ON
-cmake --build build --config Release
+cmake --build build --config Release -j$(nproc)
 ```
 
 </NewCodeBlock>
@@ -60,7 +63,7 @@ Llama.cpp 已集成 Arm KleidiAI 库，该库针对 SME、I8MM 及点积加速
 
 ```bash
 cmake -B build -DGGML_CPU_KLEIDIAI=ON
-cmake --build build --config Release
+cmake --build build --config Release -j$(nproc)
 ```
 
 </NewCodeBlock>
diff --git a/i18n/en/docusaurus-plugin-content-docs/current/common/ai/_llama_cpp.mdx b/i18n/en/docusaurus-plugin-content-docs/current/common/ai/_llama_cpp.mdx
index 2e14bbcdb..25191231f 100644
--- a/i18n/en/docusaurus-plugin-content-docs/current/common/ai/_llama_cpp.mdx
+++ b/i18n/en/docusaurus-plugin-content-docs/current/common/ai/_llama_cpp.mdx
@@ -23,7 +23,7 @@ git clone https://github.com/ggml-org/llama.cpp.git
 <NewCodeBlock tip="Device" type="device">
 
 ```bash
-sudo apt install cmake gcc g++
+sudo apt install cmake gcc g++ libcurl4-openssl-dev
 ```
 
 </NewCodeBlock>
@@ -39,15 +39,18 @@ cmake --build build --config Release -j$(nproc)
 
 </NewCodeBlock>
 
-:::info[KleidiAI]
+:::info[ARMv9]
 On devices using the ARMv9 architecture, such as [Radxa Orion O6 / O6N](/orion/o6), you can enable the `armv9-a` and KleidiAI build options for hardware-level optimization.
+
+**Please use `4aced7a` commit**.
 :::
 
 <NewCodeBlock tip="Device" type="device">
 
 ```bash
+git checkout 4aced7a
 cmake -B build -DGGML_NATIVE=OFF -DGGML_CPU_ARM_ARCH=armv9-a+i8mm+dotprod -DGGML_CPU_KLEIDIAI=ON
-cmake --build build --config Release
+cmake --build build --config Release -j$(nproc)
 ```
 
 </NewCodeBlock>
@@ -60,7 +63,7 @@ llama.cpp integrates the Arm KleidiAI library, which provides highly optimized m
 
 ```bash
 cmake -B build -DGGML_CPU_KLEIDIAI=ON
-cmake --build build --config Release
+cmake --build build --config Release -j$(nproc)
 ```
 
 </NewCodeBlock>