LLM: Modify CPU Installation Command for most examples (#11049)

* init

* refine

* refine

* refine

* modify hf-agent example

* modify all CPU model example

* remove readthedoc modify

* replace powershell with cmd

* fix repo

* fix repo

* update

* remove comment on windows code block

* update

* update

* update

* update

---------

Co-authored-by: xiangyuT <xiangyu.tian@intel.com>

README.md

@@ -110,7 +110,7 @@ See the demo of running [*Text-Generation-WebUI*](https://ipex-llm.readthedocs.i
   - LLM finetuning on Intel [GPU](python/llm/example/GPU/LLM-Finetuning), including [LoRA](python/llm/example/GPU/LLM-Finetuning/LoRA), [QLoRA](python/llm/example/GPU/LLM-Finetuning/QLoRA), [DPO](python/llm/example/GPU/LLM-Finetuning/DPO), [QA-LoRA](python/llm/example/GPU/LLM-Finetuning/QA-LoRA) and [ReLoRA](python/llm/example/GPU/LLM-Finetuning/ReLora)
   - QLoRA finetuning on Intel [CPU](python/llm/example/CPU/QLoRA-FineTuning)
 - Integration with community libraries
-  - [HuggingFace tansformers](python/llm/example/GPU/HF-Transformers-AutoModels)
+  - [HuggingFace transformers](python/llm/example/GPU/HF-Transformers-AutoModels)
   - [Standard PyTorch model](python/llm/example/GPU/PyTorch-Models)
   - [DeepSpeed-AutoTP](python/llm/example/GPU/Deepspeed-AutoTP)
   - [HuggingFace PEFT](python/llm/example/GPU/LLM-Finetuning/HF-PEFT)

docs/readthedocs/source/doc/LLM/Overview/install_cpu.md

@@ -97,4 +97,4 @@ Then for running a LLM model with IPEX-LLM optimizations (taking an `example.py`
          # e.g. for a server with 48 cores per socket
          export OMP_NUM_THREADS=48
          numactl -C 0-47 -m 0 python example.py
-```
+```

docs/readthedocs/source/index.rst

@@ -162,7 +162,7 @@ Code Examples
 
 * Integration with community libraries
 
-  * `HuggingFace tansformers <https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/HF-Transformers-AutoModels>`_
+  * `HuggingFace transformers <https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/HF-Transformers-AutoModels>`_
   * `Standard PyTorch model <https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/PyTorch-Models>`_
   * `DeepSpeed-AutoTP <https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/Deepspeed-AutoTP>`_
   * `HuggingFace PEFT <https://github.com/intel-analytics/ipex-llm/tree/main/python/llm/example/GPU/LLM-Finetuning/HF-PEFT>`_

python/llm/example/CPU/Applications/hf-agent/README.md

@@ -9,14 +9,26 @@ To run this example with IPEX-LLM, we have some recommended requirements for you
 
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+ # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install pillow # additional package required for opening images
 ```
 
+On Windows:
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install pillow
+```
+
 ### 2. Run
 ```
 python ./run_agent.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --image-path IMAGE_PATH
@@ -32,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./run_agent.py --image-path IMAGE_PATH
 ```
 

python/llm/example/CPU/Applications/streaming-llm/README.md

@@ -9,10 +9,20 @@ model = AutoModelForCausalLM.from_pretrained(model_name_or_path, load_in_4bit=Tr
 
 ## Prepare Environment
 We suggest using conda to manage environment:
+
+On Linux
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
 pip install --pre --upgrade ipex-llm[all]
 ```
 

python/llm/example/CPU/Deepspeed-AutoTP/install.sh

@@ -20,4 +20,4 @@ pip install deepspeed==0.11.1
 # 4. exclude intel deepspeed extension, which is only for XPU
 pip uninstall intel-extension-for-deepspeed
 # 5. install ipex-llm
-pip install --pre --upgrade ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/AWQ/README.md

@@ -33,16 +33,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a AWQ
 
 We suggest using conda to manage environment:
 
+On Linux
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
 pip install autoawq==0.1.8 --no-deps
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.35.0
 pip install accelerate==0.25.0
 pip install einops
 ```
+On Windows:
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install autoawq==0.1.8 --no-deps
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.35.0
+pip install accelerate==0.25.0
+pip install einops
+```
+
 **Note: For Mixtral model, please use transformers 4.36.0:**
 ```bash
 pip install transformers==4.36.0
@@ -68,7 +83,7 @@ Arguments info:
 
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
 
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/GGUF/README.md

@@ -24,19 +24,32 @@ In the example [generate.py](./generate.py), we show a basic use case to load a
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.36.0  # upgrade transformers
 ```
+
+On Windows:
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all] 
+pip install transformers==4.36.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --model <path_to_gguf_model> --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/GPTQ/README.md

@@ -8,16 +8,31 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Llama2 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.34.0
 BUILD_CUDA_EXT=0 pip install git+https://github.com/PanQiWei/AutoGPTQ.git@1de9ab6
 pip install optimum==0.14.0
 ```
 
+On Windows:
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.34.0
+set BUILD_CUDA_EXT=0
+pip install git+https://github.com/PanQiWei/AutoGPTQ.git@1de9ab6
+pip install optimum==0.14.0
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -34,7 +49,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/README.md

@@ -9,6 +9,6 @@ For OS, IPEX-LLM supports Ubuntu 20.04 or later (glibc>=2.17), CentOS 7 or later
 ## Best Known Configuration on Linux
 For better performance, it is recommended to set environment variables on Linux with the help of IPEX-LLM:
 ```bash
-pip install ipex-llm
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 source ipex-llm-init
 ```

python/llm/example/CPU/HF-Transformers-AutoModels/Model/aquila/README.md

@@ -15,11 +15,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Aqui
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/aquila2/README.md

@@ -15,11 +15,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Aqui
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/baichuan/README.md

@@ -9,12 +9,14 @@ In the example [generate.py](./generate.py), we show a basic use case for a Baic
 ### 1. Install
 We suggest using conda to manage environment:
 
+
 On Linux:
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers_stream_generator  # additional package required for Baichuan-13B-Chat to conduct generation
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/baichuan2/README.md

@@ -8,14 +8,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Baichuan model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+
 pip install transformers_stream_generator  # additional package required for Baichuan-13B-Chat to conduct generation
 ```
 
+On Windows:
+```cmd
+onda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers_stream_generator
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -32,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/bluelm/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a BlueLM model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm/README.md

@@ -15,14 +15,28 @@ In the example [generate.py](./generate.py), we show a basic use case for a Chat
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install "transformers<4.34.1"  # chatglm cannot work with transformers 4.34.1+
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install "transformers<4.34.1"
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -32,7 +46,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm2/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a ChatGLM2 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 
@@ -79,11 +92,24 @@ Inference time: xxxx s
 In the example [streamchat.py](./streamchat.py), we show a basic use case for a ChatGLM2 model to stream chat, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -108,7 +134,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 $env:PYTHONUNBUFFERED=1  # ensure stdout and stderr streams are sent straight to terminal without being first buffered
 python ./streamchat.py
 ```

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm3/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a ChatGLM3 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 
@@ -80,11 +93,24 @@ AI stands for Artificial Intelligence. It refers to the development of computer
 In the example [streamchat.py](./streamchat.py), we show a basic use case for a ChatGLM3 model to stream chat, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -109,7 +135,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 $env:PYTHONUNBUFFERED=1  # ensure stdout and stderr streams are sent straight to terminal without being first buffered
 python ./streamchat.py
 ```

python/llm/example/CPU/HF-Transformers-AutoModels/Model/codegemma/README.md

@@ -10,17 +10,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Code
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
 # install ipex-llm with 'all' option
-pip install ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # According to CodeGemma's requirement, please make sure you are using a stable version of Transformers, 4.38.1 or newer.
 pip install transformers==4.38.1
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.38.1
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -37,7 +51,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/codellama/README.md

@@ -8,14 +8,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a CodeLlama model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.34.1 # CodeLlamaTokenizer is supported in higher version of transformers
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.34.1
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -32,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/codeshell/README.md

@@ -15,11 +15,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Code
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'def print_hello_world():'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/cohere/README.md

@@ -8,12 +8,26 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a cohere model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
-pip install tansformers==4.40.0
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install transformers==4.40.0
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.40.0
 ```
 
 ### 2. Run
@@ -32,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deciLM-7b/README.md

@@ -8,14 +8,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a DeciLM-7B model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.35.2 # required by DeciLM-7B
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.35.2
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -32,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deepseek-moe/README.md

@@ -15,14 +15,28 @@ In the example [generate.py](./generate.py), we show a basic use case for a Deep
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for DeepSeek-MoE to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -35,7 +49,7 @@ You need to disable flash attention to run this model. To do this, simply replac
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deepseek/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Deepseek model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/distil-whisper/README.md

@@ -11,14 +11,28 @@ In the example [recognize.py](./recognize.py), we show a basic use case for a Di
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install datasets soundfile librosa # required by audio processing
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install datasets soundfile librosa
+```
+
 ### 2. Run
 ```
 python ./recognize.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --repo-id-or-data-path REPO_ID_OR_DATA_PATH --language LANGUAGE --chunk-length CHUNK_LENGTH --batch-size BATCH_SIZE
@@ -38,7 +52,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./recognize.py
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/dolly_v1/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Dolly v1 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/dolly_v2/README.md

@@ -8,13 +8,25 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Dolly v2 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
-
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -31,7 +43,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/falcon/README.md

@@ -9,14 +9,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Falcon model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops # additional package required for falcon-7b-instruct and falcon-40b-instruct to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. (Optional) Download Model and Replace File
 If you select the Falcon models ([tiiuae/falcon-7b-instruct](https://huggingface.co/tiiuae/falcon-7b-instruct) or [tiiuae/falcon-40b-instruct](https://huggingface.co/tiiuae/falcon-40b-instruct)), please note that their code (`modelling_RW.py`) does not support KV cache at the moment. To address issue, we have provided two updated files ([falcon-7b-instruct/modelling_RW.py](./falcon-7b-instruct/modelling_RW.py) and [falcon-40b-instruct/modelling_RW.py](./falcon-40b-instruct/modelling_RW.py)), which can be used to achieve the best performance using IPEX-LLM INT4 optimizations with KV cache support.
 After transformers 4.36, only transformer models are supported since remote code diverges from transformer model code, make sure set `trust_remote_code=False`.
@@ -66,7 +80,7 @@ Arguments info:
 
 #### 3.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/flan-t5/README.md

@@ -11,11 +11,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Flan
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -27,7 +40,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'Translate to German: My name is Arthur'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/fuyu/README.md

@@ -10,15 +10,29 @@ In the example [generate.py](./generate.py), we show a basic use case for an Fuy
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 pip install transformers==4.35 pillow # additional package required for Fuyu to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.35 pillow
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -28,7 +42,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --image-path demo.jpg
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/gemma/README.md

@@ -11,16 +11,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Gemm
 We suggest using conda to manage the Python environment:
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # According to Gemma's requirement, please make sure you are using a stable version of Transformers, 4.38.1 or newer.
 pip install transformers==4.38.1
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.38.1
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -37,7 +52,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm-xcomposer/README.md

@@ -10,16 +10,32 @@ In the example [chat.py](./chat.py), we show a basic use case for an InternLM_XC
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 pip install accelerate timm==0.4.12 sentencepiece==0.1.99 gradio==3.44.4 markdown2==2.4.10 xlsxwriter==3.1.2 einops # additional package required for InternLM_XComposer to conduct generation
 
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install accelerate timm==0.4.12 sentencepiece==0.1.99 gradio==3.44.4 markdown2==2.4.10 xlsxwriter==3.1.2 einops
+
+```
+
 ### 2. Download Model and Replace File
 If you select the InternLM_XComposer model ([internlm/internlm-xcomposer-vl-7b](https://huggingface.co/internlm/internlm-xcomposer-vl-7b)), please note that their code (`modeling_InternLM_XComposer.py`) does not support inference on CPU. To address this issue, we have provided the updated file ([internlm-xcomposer-vl-7b/modeling_InternLM_XComposer.py](./internlm-xcomposer-vl-7b/modeling_InternLM_XComposer.py), which can be used to conduct inference on CPU.
 
@@ -49,7 +65,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 3.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./chat.py --image-path demo.jpg
 ```
 More information about arguments can be found in [Arguments Info](#33-arguments-info) section. The expected output can be found in [Sample Output](#34-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a InternLM model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm2/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a InternLM2 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/llama2/README.md

@@ -8,13 +8,25 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Llama2 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
-
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -31,7 +43,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/llama3/README.md

@@ -8,13 +8,27 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Llama3 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # transformers>=4.33.0 is required for Llama3 with IPEX-LLM optimizations
+pip install transformers==4.37.0 
+```
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
 pip install transformers==4.37.0 
 ```
 
@@ -34,7 +48,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mistral/README.md

@@ -12,16 +12,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Mist
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # Refer to https://huggingface.co/mistralai/Mistral-7B-v0.1#troubleshooting, please make sure you are using a stable version of Transformers, 4.34.0 or newer.
 pip install transformers==4.34.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.34.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -31,7 +46,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mixtral/README.md

@@ -12,18 +12,30 @@ In the example [generate.py](./generate.py), we show a basic use case for a Mixt
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-# below command will install PyTorch CPU as default
-pip install torch==2.0.1 --index-url https://download.pytorch.org/whl/cpu
-pip install --pre --upgrade ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # Please make sure you are using a stable version of Transformers, 4.36.0 or newer.
 pip install transformers==4.36.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.36.0
+```
+
 ### 2. Run
 
 ```bash

python/llm/example/CPU/HF-Transformers-AutoModels/Model/moss/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a MOSS model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install "transformers<4.34"
+```
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install "transformers<4.34"
 ```
 
@@ -33,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mpt/README.md

@@ -8,14 +8,27 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for an MPT model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for mpt-7b-chat and mpt-30b-chat to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phi-1_5/README.md

@@ -15,14 +15,28 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for phi-1_5 to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -32,7 +46,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phi-2/README.md

@@ -15,13 +15,26 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for phi-2 to conduct generation
 ```
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
 
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
@@ -32,7 +45,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phi-3/README.md

@@ -15,11 +15,25 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+
+pip install transformers==4.37.0
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 
 pip install transformers==4.37.0
 ```
@@ -33,7 +47,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phixtral/README.md

@@ -15,14 +15,28 @@ In the example [generate.py](./generate.py), we show a basic use case for a phix
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for phi to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -32,7 +46,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phoenix/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Phoenix model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen-vl/README.md

@@ -10,22 +10,38 @@ In the example [chat.py](./chat.py), we show a basic use case for a Qwen-VL mode
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 pip install accelerate tiktoken einops transformers_stream_generator==0.0.4 scipy torchvision pillow tensorboard matplotlib # additional package required for Qwen-VL-Chat to conduct generation
 
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install accelerate tiktoken einops transformers_stream_generator==0.0.4 scipy torchvision pillow tensorboard matplotlib
+
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./chat.py
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen/README.md

@@ -14,14 +14,27 @@ In the example [generate.py](./generate.py), we show a basic use case for a Qwen
 
 We suggest using conda to manage environment:
 
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install tiktoken einops transformers_stream_generator  # additional package required for Qwen-7B-Chat to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install tiktoken einops transformers_stream_generator
+```
+
 ### 2. Run
 
 The minimum Qwen model version currently supported by IPEX-LLM is the version on November 30, 2023.
@@ -44,7 +57,7 @@ Arguments info:
 
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
 
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen1.5/README.md

@@ -9,11 +9,15 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Qwen model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.37.0 # install the transformers which support Qwen2
 
 # only for Qwen1.5-MoE-A2.7B
@@ -21,6 +25,20 @@ pip install transformers==4.40.0
 pip install trl==0.8.1
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.37.0
+
+REM For Qwen1.5-MoE-A2.7B
+pip install transformers==4.40.0
+pip install trl==0.8.1
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -37,7 +55,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/redpajama/README.md

@@ -9,11 +9,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a RedPajama model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -32,7 +45,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/replit/README.md

@@ -10,11 +10,25 @@ In the example [generate.py](./generate.py), we show a basic use case for an Rep
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install "transformers<4.35"
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install "transformers<4.35"
 ```
 
@@ -23,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'def print_hello_world():'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/skywork/README.md

@@ -8,13 +8,25 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Skywork model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
-
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -31,7 +43,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/solar/README.md

@@ -8,14 +8,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a SOLAR model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.35.2 # required by SOLAR
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.35.2
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -32,7 +46,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/stablelm/README.md

@@ -10,16 +10,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Stab
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # Refer to https://huggingface.co/stabilityai/stablelm-zephyr-3b/blob/8b471c751c0e78cb46cf9f47738dd0eb45392071/config.json#L21, please make sure you are using a stable version of Transformers, 4.38.0 or newer.
 pip install transformers==4.38.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.38.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -29,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/starcoder/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for an StarCoder model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/vicuna/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a Vicuna model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/whisper/readme.md

@@ -9,14 +9,28 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [recognize.py](./recognize.py), we show a basic use case for a Whisper model to conduct transcription using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install datasets soundfile librosa # required by audio processing
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install datasets soundfile librosa
+```
+
 ### 2. Run
 ```
 python ./recognize.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --repo-id-or-data-path REPO_ID_OR_DATA_PATH --language LANGUAGE
@@ -34,7 +48,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./recognize.py 
 ```
 
@@ -65,11 +79,25 @@ Inference time: xxxx s
 In the example [long-segment-recognize.py](./long-segment-recognize.py), we show a basic use case for a Whisper model to conduct transcription using `pipeline()` API for long audio input, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install datasets soundfile librosa # required by audio processing
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install datasets soundfile librosa # required by audio processing
 ```
 
@@ -92,7 +120,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 # Long Segment Recognize
 python ./long-segment-recognize.py --audio-file /PATH/TO/AUDIO_FILE
 ```

python/llm/example/CPU/HF-Transformers-AutoModels/Model/wizardcoder-python/README.md

@@ -8,11 +8,24 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a WizardCoder-Python model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -31,7 +44,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/yi/README.md

@@ -10,14 +10,28 @@ In the example [generate.py](./generate.py), we show a basic use case for an Yi
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops # additional package required for Yi-6B to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -34,7 +48,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/yuan2/README.md

@@ -12,15 +12,30 @@ In the example [generate.py](./generate.py), we show a basic use case for an Yua
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops # additional package required for Yuan2 to conduct generation
 pip install pandas # additional package required for Yuan2 to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+pip install pandas
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -37,7 +52,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/ziya/README.md

@@ -15,14 +15,28 @@ In the example [generate.py](./generate.py), we show a basic use case for a Ziya
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops  # additional package required for Ziya to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -32,7 +46,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'def quick_sort(arr):\n'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/HF-Transformers-AutoModels/More-Data-Types/README.md

@@ -8,7 +8,7 @@ We suggest using conda to manage environment:
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 ```
 
 ## Run Example

python/llm/example/CPU/HF-Transformers-AutoModels/Save-Load/README.md

@@ -8,7 +8,7 @@ We suggest using conda to manage environment:
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 ```
 
 ## Run Example

python/llm/example/CPU/ModelScope-Models/README.md

@@ -9,15 +9,29 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a ChatGLM3 model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
+# install ipex-llm with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 # Refer to https://github.com/modelscope/modelscope/issues/765, please make sure you are using 1.11.0 version
 pip install modelscope==1.11.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install modelscope==1.11.0
+```
+
 ### 2. Run
 ```
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
@@ -34,7 +48,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/Native-Models/README.md

@@ -6,7 +6,20 @@ In this example, we show a pipeline to convert a large language model to IPEX-LL
 
 ## Prepare Environment
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 

python/llm/example/CPU/PyTorch-Models/Model/aquila2/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Aqui
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/bark/README.md

@@ -10,11 +10,26 @@ In the example [synthesize_speech.py](./synthesize_speech.py), we show a basic u
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install TTS scipy
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install TTS scipy
 ```
 
@@ -34,7 +49,7 @@ After setting up the Python environment and downloading Bark model, you could ru
 
 #### 3.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 # make sure `--model-path` corresponds to the local folder of downloaded model
 python ./synthesize_speech.py --model-path 'bark/' --text "This is an example text for synthesize speech."
 ```

python/llm/example/CPU/PyTorch-Models/Model/bert/README.md

@@ -10,19 +10,31 @@ In the example [extract_feature.py](./extract_feature.py), we show a basic use c
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+```
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./extract_feature.py --text 'This is an example text for feature extraction.'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section.

python/llm/example/CPU/PyTorch-Models/Model/bluelm/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Blue
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/chatglm/README.md

@@ -10,12 +10,27 @@ In the example [generate.py](./generate.py), we show a basic use case for a Chat
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
-pip install "transformers<4.34.1"  # chatglm cannot work with transformers 4.34.1+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install "transformers<4.34.1"
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install "transformers<4.34.1"
 ```
 
 ### 2. Run
@@ -23,7 +38,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/chatglm3/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Chat
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'AI是什么？'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/codegemma/README.md

@@ -10,23 +10,36 @@ In the example [generate.py](./generate.py), we show a basic use case for a Code
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-# install ipex-llm with 'all' option
-pip install --pre --upgrade ipex-llm[all]
-
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 # According to CodeGemma's requirement, please make sure you are using a stable version of Transformers, 4.38.1 or newer.
 pip install transformers==4.38.1
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.38.1
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/codellama/README.md

@@ -10,20 +10,35 @@ In the example [generate.py](./generate.py), we show a basic use case for a Code
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.34.1 # CodeLlamaTokenizer is supported in higher version of transformers
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.34.1
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'def print_hello_world():'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/codeshell/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Code
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'def print_hello_world():'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/cohere/README.md

@@ -8,12 +8,27 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [generate.py](./generate.py), we show a basic use case for a cohere model to predict the next N tokens using `generate()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install transformers==4.40.0
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install ipex-llm with 'all' option
-pip install tansformers==4.40.0
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.40.0
 ```
 
 ### 2. Run
@@ -32,7 +47,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py 
 ```
 

python/llm/example/CPU/PyTorch-Models/Model/deciLM-7b/README.md

@@ -10,20 +10,34 @@ In the example [generate.py](./generate.py), we show a basic use case for a Deci
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.35.2 # required by DeciLM-7B
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.35.2
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/deepseek-moe/README.md

@@ -10,11 +10,25 @@ In the example [generate.py](./generate.py), we show a basic use case for a deep
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install einops 
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install einops 
 ```
 
@@ -23,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/deepseek/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Deep
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/distil-whisper/README.md

@@ -11,12 +11,26 @@ In the example [recognize.py](./recognize.py), we show a basic use case for a Di
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
+conda create -n llm python=3.11 # recommend to use Python 3.11
+conda activate llm
+
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install datasets soundfile librosa # required by audio processing
+```
+
+On Windows:
+
+```cmd
 conda create -n llm python=3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
-pip install datasets soundfile librosa # required by audio processing
+pip install --pre --upgrade ipex-llm[all]
+pip install datasets soundfile librosa
 ```
 
 ### 2. Run
@@ -28,7 +42,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./recognize.py
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/flan-t5/README.md

@@ -11,11 +11,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Flan
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -27,7 +40,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'Translate to German: My name is Arthur'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/fuyu/README.md

@@ -10,15 +10,30 @@ In the example [generate.py](./generate.py), we show a basic use case for an Fuy
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 pip install transformers==4.35 pillow # additional package required for Fuyu to conduct generation
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.35 pillow
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -28,7 +43,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --image-path demo.jpg
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/internlm-xcomposer/README.md

@@ -10,14 +10,28 @@ In the example [chat.py](./chat.py), we show a basic use case for an InternLM_XC
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 pip install accelerate timm==0.4.12 sentencepiece==0.1.99 gradio==3.44.4 markdown2==2.4.10 xlsxwriter==3.1.2 einops # additional package required for InternLM_XComposer to conduct generation
+```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install accelerate timm==0.4.12 sentencepiece==0.1.99 gradio==3.44.4 markdown2==2.4.10 xlsxwriter==3.1.2 einops
 ```
 
 ### 2. Download Model and Replace File
@@ -49,7 +63,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 3.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./chat.py --image-path demo.jpg
 ```
 More information about arguments can be found in [Arguments Info](#33-arguments-info) section. The expected output can be found in [Sample Output](#34-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/internlm2/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Inte
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/llama2/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Llam
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/llama3/README.md

@@ -10,22 +10,37 @@ In the example [generate.py](./generate.py), we show a basic use case for a Llam
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # transformers>=4.33.0 is required for Llama3 with IPEX-LLM optimizations
 pip install transformers==4.37.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.37.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/llava/README.md

@@ -11,11 +11,15 @@ In the example [generate.py](./generate.py), we show a basic use case for a LLaV
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops # install dependencies required by llava
 pip install transformers==4.36.2
 
@@ -25,6 +29,22 @@ cd LLaVA # change the working directory to the LLaVA folder
 git checkout tags/v1.2.0 -b 1.2.0 # Get the branch which is compatible with transformers 4.36
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+pip install transformers==4.36.2
+
+git clone https://github.com/haotian-liu/LLaVA.git
+copy generate.py .\LLaVA\
+cd LLaVA
+git checkout tags/v1.2.0 -b 1.2.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -34,7 +54,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --image-path-or-url 'https://llava-vl.github.io/static/images/monalisa.jpg'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/mamba/README.md

@@ -10,20 +10,34 @@ In the example [generate.py](./generate.py), we show a basic use case for a Mamb
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install einops # package required by Mamba
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install einops
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 

python/llm/example/CPU/PyTorch-Models/Model/meta-llama/README.md

@@ -9,6 +9,9 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [example_chat_completion.py](./example_chat_completion.py), we show a basic use case for a Llama model to engage in a conversation with an AI assistant using `chat_completion` API, with IPEX-LLM INT4 optimizations. The process for [example_text_completion.py](./example_text_completion.py) is similar.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
@@ -19,8 +22,22 @@ cd llama/
 git apply < ../cpu.patch # apply cpu version patch
 pip install -e .
 
-cd -
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+git clone https://github.com/facebookresearch/llama.git
+cd llama/
+git apply < ../cpu.patch
+pip install -e .
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -46,7 +63,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 torchrun --nproc-per-node 1 example_chat_completion.py --ckpt_dir llama-2-7b-chat/ --tokenizer_path tokenizer.model --max_seq_len 64 --max_batch_size 1 --backend cpu
 ```
 

python/llm/example/CPU/PyTorch-Models/Model/mistral/README.md

@@ -12,22 +12,35 @@ In the example [generate.py](./generate.py), we show a basic use case for a Mist
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
-
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 # Refer to https://huggingface.co/mistralai/Mistral-7B-v0.1#troubleshooting, please make sure you are using a stable version of Transformers, 4.34.0 or newer.
 pip install transformers==4.34.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.34.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/mixtral/README.md

@@ -12,18 +12,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Mixt
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-# below command will install PyTorch CPU as default
-pip install torch==2.0.1 --index-url https://download.pytorch.org/whl/cpu
-pip install --pre --upgrade ipex-llm[all]
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+
 
 # Please make sure you are using a stable version of Transformers, 4.36.0 or newer.
 pip install transformers==4.36.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.36.0
+```
+
 ### 2. Run
 
 ```bash

python/llm/example/CPU/PyTorch-Models/Model/openai-whisper/readme.md

@@ -9,15 +9,30 @@ To run these examples with IPEX-LLM, we have some recommended requirements for y
 In the example [recognize.py](./recognize.py), we show a basic use case for a Whisper model to conduct transcription using `transcribe()` API, with IPEX-LLM INT4 optimizations.
 ### 1. Install
 We suggest using conda to manage environment:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11
 conda activate llm
 
-pip install ipex-llm[all] # install ipex-llm with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install -U openai-whisper
 pip install librosa # required by audio processing
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install -U openai-whisper
+pip install librosa
+```
+
 ### 2. Run
 ```
 python ./recognize.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --repo-id-or-data-path REPO_ID_OR_DATA_PATH --language LANGUAGE
@@ -35,7 +50,7 @@ Arguments info:
 
 #### 2.1 Client
 On client Windows machine, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./recognize.py --audio-file /PATH/TO/AUDIO_FILE
 ```
 

python/llm/example/CPU/PyTorch-Models/Model/phi-1_5/README.md

@@ -10,11 +10,25 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install einops 
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install einops 
 ```
 
@@ -23,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/phi-2/README.md

@@ -10,11 +10,25 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install einops 
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install einops 
 ```
 
@@ -23,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/phi-3/README.md

@@ -15,11 +15,26 @@ In the example [generate.py](./generate.py), we show a basic use case for a phi-
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+
+pip install transformers==4.37.0
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 
 pip install transformers==4.37.0
 ```
@@ -29,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/phixtral/README.md

@@ -10,11 +10,25 @@ In the example [generate.py](./generate.py), we show a basic use case for a phix
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+pip install einops 
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 pip install einops 
 ```
 
@@ -23,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/qwen-vl/README.md

@@ -10,14 +10,26 @@ In the example [chat.py](./chat.py), we show a basic use case for a Qwen-VL mode
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
-
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install accelerate tiktoken einops transformers_stream_generator==0.0.4 scipy torchvision pillow tensorboard matplotlib # additional package required for Qwen-VL-Chat to conduct generation
+```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install accelerate tiktoken einops transformers_stream_generator==0.0.4 scipy torchvision pillow tensorboard matplotlib
 ```
 
 ### 2. Run
@@ -25,7 +37,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./chat.py
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/qwen1.5/README.md

@@ -10,11 +10,15 @@ In the example [generate.py](./generate.py), we show a basic use case for a Qwen
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.37.0 # install transformers which supports Qwen2
 
 # only for Qwen1.5-MoE-A2.7B
@@ -22,12 +26,26 @@ pip install transformers==4.40.0
 pip install trl==0.8.1
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.37.0
+
+REM for Qwen1.5-MoE-A2.7B
+pip install transformers==4.40.0
+pip install trl==0.8.1
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/skywork/README.md

@@ -10,11 +10,24 @@ In the example [generate.py](./generate.py), we show a basic use case for a Skyw
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
+```
+
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
 ```
 
 ### 2. Run
@@ -22,7 +35,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/solar/README.md

@@ -10,20 +10,34 @@ In the example [generate.py](./generate.py), we show a basic use case for a SOLA
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 pip install transformers==4.35.2 # required by SOLAR
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+pip install transformers==4.35.2
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.

python/llm/example/CPU/PyTorch-Models/Model/stablelm/README.md

@@ -10,16 +10,31 @@ In the example [generate.py](./generate.py), we show a basic use case for a Stab
 We suggest using conda to manage the Python environment. For more information about conda installation, please refer to [here](https://docs.conda.io/en/latest/miniconda.html#).
 
 After installing conda, create a Python environment for IPEX-LLM:
+
+On Linux:
+
 ```bash
 conda create -n llm python=3.11 # recommend to use Python 3.11
 conda activate llm
 
-pip install --pre --upgrade ipex-llm[all] # install the latest ipex-llm nightly build with 'all' option
+# install the latest ipex-llm nightly build with 'all' option
+pip install --pre --upgrade ipex-llm[all] --extra-index-url https://download.pytorch.org/whl/cpu
 
 # Refer to https://huggingface.co/stabilityai/stablelm-zephyr-3b/blob/8b471c751c0e78cb46cf9f47738dd0eb45392071/config.json#L21, please make sure you are using a stable version of Transformers, 4.38.0 or newer.
 pip install transformers==4.38.0
 ```
 
+On Windows:
+
+```cmd
+conda create -n llm python=3.11
+conda activate llm
+
+pip install --pre --upgrade ipex-llm[all]
+
+pip install transformers==4.38.0
+```
+
 ### 2. Run
 After setting up the Python environment, you could run the example by following steps.
 
@@ -29,7 +44,7 @@ After setting up the Python environment, you could run the example by following
 
 #### 2.1 Client
 On client Windows machines, it is recommended to run directly with full utilization of all cores:
-```powershell
+```cmd
 python ./generate.py --prompt 'What is AI?'
 ```
 More information about arguments can be found in [Arguments Info](#23-arguments-info) section. The expected output can be found in [Sample Output](#24-sample-output) section.