Refactor bigdl.llm to ipex_llm (#24)

* Rename bigdl/llm to ipex_llm

* rm python/llm/src/bigdl

* from bigdl.llm to from ipex_llm

README.md

@@ -86,7 +86,7 @@ You may apply INT4 optimizations to any Hugging Face *Transformers* models as fo
 
 ```python
 #load Hugging Face Transformers model with INT4 optimizations
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
 #run the optimized model on CPU
@@ -113,7 +113,7 @@ You may apply INT4 optimizations to any Hugging Face *Transformers* models as fo
 
 ```python
 #load Hugging Face Transformers model with INT4 optimizations
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
 #run the optimized model on Intel GPU

docker/llm/README.md

@@ -223,7 +223,7 @@ This controller manages the distributed workers.
 
 ##### Launch the model worker(s)
 ```bash
-python3 -m bigdl.llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device cpu
+python3 -m ipex_llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device cpu
 ```
 Wait until the process finishes loading the model and you see "Uvicorn running on ...". The model worker will register itself to the controller.
 
@@ -252,7 +252,7 @@ python3 -m fastchat.serve.controller
 Then, launch the model worker(s):
 
 ```bash
-python3 -m bigdl.llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device cpu
+python3 -m ipex_llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device cpu
 ```
 
 Finally, launch the RESTful API server
@@ -319,7 +319,7 @@ This controller manages the distributed workers.
 
 ##### Launch the model worker(s)
 ```bash
-python3 -m bigdl.llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device xpu
+python3 -m ipex_llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device xpu
 ```
 Wait until the process finishes loading the model and you see "Uvicorn running on ...". The model worker will register itself to the controller.
 
@@ -346,7 +346,7 @@ python3 -m fastchat.serve.controller
 Then, launch the model worker(s):
 
 ```bash
-python3 -m bigdl.llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device xpu
+python3 -m ipex_llm.serving.model_worker --model-path lmsys/vicuna-7b-v1.3 --device xpu
 ```
 
 Finally, launch the RESTful API server

docker/llm/inference/xpu/docker/chat.py

@@ -23,7 +23,7 @@ from transformers import TextIteratorStreamer
 from transformers.tools.agents import StopSequenceCriteria
 from transformers.generation.stopping_criteria import StoppingCriteriaList
 from colorama import Fore
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 SYSTEM_PROMPT = "A chat between a curious human <human> and an artificial intelligence assistant <bot>.\
 The assistant gives helpful, detailed, and polite answers to the human's questions."
 HUMAN_ID = "<human>"

docker/llm/serving/cpu/docker/entrypoint.sh

@@ -135,9 +135,9 @@ else
   done
 
   if [ "$worker_type" == "model_worker" ]; then
-      worker_type="bigdl.llm.serving.model_worker"
+      worker_type="ipex_llm.serving.model_worker"
   elif [ "$worker_type" == "vllm_worker" ]; then
-      worker_type="bigdl.llm.serving.vllm_worker"
+      worker_type="ipex_llm.serving.vllm_worker"
   fi
 
   if [[ -n $CONTROLLER_HOST ]]; then
@@ -220,9 +220,9 @@ else
     echo "Worker type: $worker_type"
     echo "Worker address: $worker_address"
     echo "Controller address: $controller_address"
-    if [ "$worker_type" == "bigdl.llm.serving.model_worker" ]; then
+    if [ "$worker_type" == "ipex_llm.serving.model_worker" ]; then
       python3 -m "$worker_type" --model-path $model_path --device cpu --host $worker_host --port $worker_port --worker-address $worker_address --controller-address $controller_address --stream-interval $stream_interval
-    elif [ "$worker_type" == "bigdl.llm.serving.vllm_worker" ]; then
+    elif [ "$worker_type" == "ipex_llm.serving.vllm_worker" ]; then
       python3 -m "$worker_type" --model-path $model_path --device cpu --host $worker_host --port $worker_port --worker-address $worker_address --controller-address $controller_address
     fi
   fi

docker/llm/serving/cpu/docker/model_adapter.py.patch

@@ -9,7 +9,7 @@
          generation_config = GenerationConfig.from_pretrained(
              model_path, trust_remote_code=True
          )
-+        from bigdl.llm.transformers import AutoModelForCausalLM
++        from ipex_llm.transformers import AutoModelForCausalLM
          model = AutoModelForCausalLM.from_pretrained(
              model_path,
              config=config,

docker/llm/serving/xpu/docker/entrypoint.sh

@@ -66,9 +66,9 @@ else
     done
     
     if [ "$worker_type" == "model_worker" ]; then
-        worker_type="bigdl.llm.serving.model_worker"
+        worker_type="ipex_llm.serving.model_worker"
     elif [ "$worker_type" == "vllm_worker" ]; then
-        worker_type="bigdl.llm.serving.vllm_worker"
+        worker_type="ipex_llm.serving.vllm_worker"
     fi
     
     if [[ -n $CONTROLLER_HOST ]]; then
@@ -127,9 +127,9 @@ else
         echo "Worker address: $worker_address"
         echo "Controller address: $controller_address"
         
-        if [ "$worker_type" == "bigdl.llm.serving.model_worker" ]; then
+        if [ "$worker_type" == "ipex_llm.serving.model_worker" ]; then
             python3 -m "$worker_type" --model-path $model_path --device xpu --host $worker_host --port $worker_port --worker-address $worker_address --controller-address $controller_address --stream-interval $stream_interval
-        elif [ "$worker_type" == "bigdl.llm.serving.vllm_worker" ]; then
+        elif [ "$worker_type" == "ipex_llm.serving.vllm_worker" ]; then
             python3 -m "$worker_type" --model-path $model_path --device xpu --host $worker_host --port $worker_port --worker-address $worker_address --controller-address $controller_address
         fi
     fi

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/finetune.md

@@ -21,7 +21,7 @@ To help you better understand the finetuning process, here we use model [Llama-2
 First, load model using `transformers`-style API and **set it to `to('xpu')`**. We specify `load_in_low_bit="nf4"` here to apply 4-bit NormalFloat optimization. According to the [QLoRA paper](https://arxiv.org/pdf/2305.14314.pdf), using `"nf4"` could yield better model quality than `"int4"`.
 
 ```python
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf",
                                              load_in_low_bit="nf4",
@@ -33,14 +33,14 @@ model = model.to('xpu')
 
 Then, we have to apply some preprocessing to the model to prepare it for training.
 ```python
-from bigdl.llm.transformers.qlora import prepare_model_for_kbit_training
+from ipex_llm.transformers.qlora import prepare_model_for_kbit_training
 model.gradient_checkpointing_enable()
 model = prepare_model_for_kbit_training(model)
 ```
 
 Next, we can obtain a Peft model from the optimized model and a configuration object containing the parameters as follows:
 ```python
-from bigdl.llm.transformers.qlora import get_peft_model
+from ipex_llm.transformers.qlora import get_peft_model
 from peft import LoraConfig
 config = LoraConfig(r=8, 
                     lora_alpha=32, 
@@ -54,7 +54,7 @@ model = get_peft_model(model, config)
 ```eval_rst
 .. important::
 
-   Instead of ``from peft import prepare_model_for_kbit_training, get_peft_model`` as we did for regular QLoRA using bitandbytes and cuda, we import them from ``bigdl.llm.transformers.qlora`` here to get a BigDL-LLM compatible Peft model. And the rest is just the same as regular LoRA finetuning process using ``peft``.
+   Instead of ``from peft import prepare_model_for_kbit_training, get_peft_model`` as we did for regular QLoRA using bitandbytes and cuda, we import them from ``ipex_llm.transformers.qlora`` here to get a BigDL-LLM compatible Peft model. And the rest is just the same as regular LoRA finetuning process using ``peft``.
 ```
 
 ```eval_rst

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/hugging_face_format.md

@@ -5,7 +5,7 @@ You may apply INT4 optimizations to any Hugging Face *Transformers* models as fo
 
 ```python
 # load Hugging Face Transformers model with INT4 optimizations
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 ```

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/inference_on_gpu.md

@@ -29,7 +29,7 @@ You could choose to use [PyTorch API](./optimize_model.html) or [`transformers`-
 
          # Take Llama-2-7b-chat-hf as an example
          from transformers import LlamaForCausalLM
-         from bigdl.llm import optimize_model
+         from ipex_llm import optimize_model
 
          model = LlamaForCausalLM.from_pretrained('meta-llama/Llama-2-7b-chat-hf', torch_dtype='auto', low_cpu_mem_usage=True)
          model = optimize_model(model) # With only one line to enable BigDL-LLM INT4 optimization
@@ -40,14 +40,14 @@ You could choose to use [PyTorch API](./optimize_model.html) or [`transformers`-
 
          When running LLMs on Intel iGPUs for Windows users, we recommend setting ``cpu_embedding=True`` in the ``optimize_model`` function. This will allow the memory-intensive embedding layer to utilize the CPU instead of iGPU.
          
-         See the `API doc <../../../PythonAPI/LLM/optimize.html#bigdl.llm.optimize_model>`_ for ``optimize_model`` to find more information.
+         See the `API doc <../../../PythonAPI/LLM/optimize.html#ipex_llm.optimize_model>`_ for ``optimize_model`` to find more information.
 
       Especially, if you have saved the optimized model following setps `here <./optimize_model.html#save>`_, the loading process on Intel GPUs maybe as follows:
 
       .. code-block:: python
 
          from transformers import LlamaForCausalLM
-         from bigdl.llm.optimize import low_memory_init, load_low_bit
+         from ipex_llm.optimize import low_memory_init, load_low_bit
 
          saved_dir='./llama-2-bigdl-llm-4-bit'
          with low_memory_init(): # Fast and low cost by loading model on meta device
@@ -65,7 +65,7 @@ You could choose to use [PyTorch API](./optimize_model.html) or [`transformers`-
       .. code-block:: python
 
          # Take Llama-2-7b-chat-hf as an example
-         from bigdl.llm.transformers import AutoModelForCausalLM
+         from ipex_llm.transformers import AutoModelForCausalLM
 
          # Load model in 4 bit, which convert the relevant layers in the model into INT4 format
          model = AutoModelForCausalLM.from_pretrained('meta-llama/Llama-2-7b-chat-hf', load_in_4bit=True)
@@ -82,7 +82,7 @@ You could choose to use [PyTorch API](./optimize_model.html) or [`transformers`-
 
       .. code-block:: python
 
-         from bigdl.llm.transformers import AutoModelForCausalLM
+         from ipex_llm.transformers import AutoModelForCausalLM
 
          saved_dir='./llama-2-bigdl-llm-4-bit'
          model = AutoModelForCausalLM.load_low_bit(saved_dir) # Load the optimized model

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/langchain_api.md

@@ -7,8 +7,8 @@ You may run the models using the LangChain API in `bigdl-llm`.
 You may run any Hugging Face *Transformers* model (with INT4 optimiztions applied) using the LangChain API as follows:
 
 ```python
-from bigdl.llm.langchain.llms import TransformersLLM
+from ipex_llm.langchain.llms import TransformersLLM
-from bigdl.llm.langchain.embeddings import TransformersEmbeddings
+from ipex_llm.langchain.embeddings import TransformersEmbeddings
 from langchain.chains.question_answering import load_qa_chain
 
 embeddings = TransformersEmbeddings.from_model_id(model_id=model_path)
@@ -37,8 +37,8 @@ You may also convert Hugging Face *Transformers* models into native INT4 format,
 ```
 
 ```python
-from bigdl.llm.langchain.llms import LlamaLLM
+from ipex_llm.langchain.llms import LlamaLLM
-from bigdl.llm.langchain.embeddings import LlamaEmbeddings
+from ipex_llm.langchain.embeddings import LlamaEmbeddings
 from langchain.chains.question_answering import load_qa_chain
 
 # switch to ChatGLMEmbeddings/GptneoxEmbeddings/BloomEmbeddings/StarcoderEmbeddings to load other models

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/native_format.md

@@ -10,13 +10,13 @@ You may also convert Hugging Face *Transformers* models into native INT4 format
 
 ```python
 # convert the model
-from bigdl.llm import llm_convert
+from ipex_llm import llm_convert
 bigdl_llm_path = llm_convert(model='/path/to/model/',
        outfile='/path/to/output/', outtype='int4', model_family="llama")
 
 # load the converted model
 # switch to ChatGLMForCausalLM/GptneoxForCausalLM/BloomForCausalLM/StarcoderForCausalLM to load other models
-from bigdl.llm.transformers import LlamaForCausalLM
+from ipex_llm.transformers import LlamaForCausalLM
 llm = LlamaForCausalLM.from_pretrained("/path/to/output/model.bin", native=True, ...)
 
 # run the converted model

docs/readthedocs/source/doc/LLM/Overview/KeyFeatures/optimize_model.md

@@ -14,7 +14,7 @@ model = LlamaForCausalLM.from_pretrained('meta-llama/Llama-2-7b-chat-hf', torch_
 
 Then, just need to call `optimize_model` to optimize the loaded model and INT4 optimization is applied on model by default: 
 ```python
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 
 # With only one line to enable BigDL-LLM INT4 optimization
 model = optimize_model(model)
@@ -31,7 +31,7 @@ Currently, ``low_bit`` supports options 'sym_int4', 'asym_int4', 'sym_int5', 'as
 You may apply symmetric INT8 optimization as follows:
 
 ```python
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 
 # Apply symmetric INT8 optimization
 model = optimize_model(model, low_bit="sym_int8")
@@ -51,7 +51,7 @@ model.save_low_bit(saved_dir)
 
 We recommend to use the context manager `low_memory_init` to quickly initiate a model instance with low cost, and then use `load_low_bit` to load the optimized low-bit model as follows:
 ```python
-from bigdl.llm.optimize import low_memory_init, load_low_bit
+from ipex_llm.optimize import low_memory_init, load_low_bit
 with low_memory_init(): # Fast and low cost by loading model on meta device
    model = LlamaForCausalLM.from_pretrained(saved_dir,
                                             torch_dtype="auto",

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

@@ -11,7 +11,7 @@ Here, let's take a relatively small LLM model, i.e [open_llama_3b_v2](https://hu
 Simply use one-line `transformers`-style API in `bigdl-llm` to load `open_llama_3b_v2` with INT4 optimization (by specifying `load_in_4bit=True`) as follows:
 
 ```python
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 model = AutoModelForCausalLM.from_pretrained(pretrained_model_name_or_path="openlm-research/open_llama_3b_v2",
                                              load_in_4bit=True)

docs/readthedocs/source/doc/LLM/Quickstart/install_linux_gpu.md

@@ -112,7 +112,7 @@ Install the Miniconda as follows if you don't have conda installed on your machi
 
   python
 
-  > from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+  > from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
   ```
 
   > <img src="https://llm-assets.readthedocs.io/en/latest/_images/verify_bigdl_import.png" alt="image-20240221102252562" width=100%; />
@@ -170,7 +170,7 @@ Now let's play with a real LLM. We'll be using the [phi-1.5](https://huggingface
    ```python
    # Copy/Paste the contents to a new file demo.py
    import torch
-   from bigdl.llm.transformers import AutoModelForCausalLM
+   from ipex_llm.transformers import AutoModelForCausalLM
    from transformers import AutoTokenizer, GenerationConfig
    generation_config = GenerationConfig(use_cache = True)
    

docs/readthedocs/source/doc/LLM/Quickstart/install_windows_gpu.md

@@ -130,7 +130,7 @@ You can verify if `bigdl-llm` is successfully installed by simply running a few
 * Step 5: Copy following code to Anaconda prompt **line by line** and press Enter **after copying each line**.
   ```python
   import torch 
-  from bigdl.llm.transformers import AutoModel,AutoModelForCausalLM    
+  from ipex_llm.transformers import AutoModel,AutoModelForCausalLM    
   tensor_1 = torch.randn(1, 1, 40, 128).to('xpu') 
   tensor_2 = torch.randn(1, 1, 128, 40).to('xpu') 
   print(torch.matmul(tensor_1, tensor_2).size()) 
@@ -200,7 +200,7 @@ Now let's play with a real LLM. We'll be using the [Qwen-1.8B-Chat](https://hugg
   
            # Copy/Paste the contents to a new file demo.py
            import torch
-           from bigdl.llm.transformers import AutoModelForCausalLM
+           from ipex_llm.transformers import AutoModelForCausalLM
            from transformers import AutoTokenizer, GenerationConfig
            generation_config = GenerationConfig(use_cache=True)
            
@@ -260,7 +260,7 @@ Now let's play with a real LLM. We'll be using the [Qwen-1.8B-Chat](https://hugg
 
            # Copy/Paste the contents to a new file demo.py
            import torch
-           from bigdl.llm.transformers import AutoModelForCausalLM
+           from ipex_llm.transformers import AutoModelForCausalLM
            from transformers import GenerationConfig
            from modelscope import AutoTokenizer
            generation_config = GenerationConfig(use_cache=True)

docs/readthedocs/source/doc/PythonAPI/LLM/langchain.rst

@@ -13,7 +13,7 @@ BigDL-LLM provides ``TransformersLLM`` and ``TransformersPipelineLLM``, which im
 
     .. tab:: AutoModel
 
-        .. automodule:: bigdl.llm.langchain.llms.transformersllm
+        .. automodule:: ipex_llm.langchain.llms.transformersllm
             :members:
             :undoc-members:
             :show-inheritance:
@@ -21,7 +21,7 @@ BigDL-LLM provides ``TransformersLLM`` and ``TransformersPipelineLLM``, which im
 
     .. tab:: pipeline
 
-        .. automodule:: bigdl.llm.langchain.llms.transformerspipelinellm
+        .. automodule:: ipex_llm.langchain.llms.transformerspipelinellm
             :members:
             :undoc-members:
             :show-inheritance:
@@ -37,7 +37,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Llama
 
-        .. autoclass:: bigdl.llm.langchain.llms.bigdlllm.LlamaLLM
+        .. autoclass:: ipex_llm.langchain.llms.bigdlllm.LlamaLLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -49,7 +49,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: ChatGLM
 
-        .. autoclass:: bigdl.llm.langchain.llms.bigdlllm.ChatGLMLLM
+        .. autoclass:: ipex_llm.langchain.llms.bigdlllm.ChatGLMLLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -61,7 +61,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Bloom
 
-        .. autoclass:: bigdl.llm.langchain.llms.bigdlllm.BloomLLM
+        .. autoclass:: ipex_llm.langchain.llms.bigdlllm.BloomLLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -73,7 +73,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Gptneox
 
-        .. autoclass:: bigdl.llm.langchain.llms.bigdlllm.GptneoxLLM
+        .. autoclass:: ipex_llm.langchain.llms.bigdlllm.GptneoxLLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -85,7 +85,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Starcoder
 
-        .. autoclass:: bigdl.llm.langchain.llms.bigdlllm.StarcoderLLM
+        .. autoclass:: ipex_llm.langchain.llms.bigdlllm.StarcoderLLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -102,7 +102,7 @@ Embeddings Wrapper of LangChain
 Hugging Face ``transformers`` AutoModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. automodule:: bigdl.llm.langchain.embeddings.transformersembeddings
+.. automodule:: ipex_llm.langchain.embeddings.transformersembeddings
     :members:
     :undoc-members:
     :show-inheritance:
@@ -117,7 +117,7 @@ For ``llama``/``bloom``/``gptneox``/``starcoder`` model families, you could also
 
     .. tab:: Llama
 
-        .. autoclass:: bigdl.llm.langchain.embeddings.bigdlllm.LlamaEmbeddings
+        .. autoclass:: ipex_llm.langchain.embeddings.bigdlllm.LlamaEmbeddings
             :members:
             :undoc-members:
             :show-inheritance:
@@ -129,7 +129,7 @@ For ``llama``/``bloom``/``gptneox``/``starcoder`` model families, you could also
 
     .. tab:: Bloom
 
-        .. autoclass:: bigdl.llm.langchain.embeddings.bigdlllm.BloomEmbeddings
+        .. autoclass:: ipex_llm.langchain.embeddings.bigdlllm.BloomEmbeddings
             :members:
             :undoc-members:
             :show-inheritance:
@@ -141,7 +141,7 @@ For ``llama``/``bloom``/``gptneox``/``starcoder`` model families, you could also
 
     .. tab:: Gptneox
 
-        .. autoclass:: bigdl.llm.langchain.embeddings.bigdlllm.GptneoxEmbeddings
+        .. autoclass:: ipex_llm.langchain.embeddings.bigdlllm.GptneoxEmbeddings
             :members:
             :undoc-members:
             :show-inheritance:
@@ -153,7 +153,7 @@ For ``llama``/``bloom``/``gptneox``/``starcoder`` model families, you could also
 
     .. tab:: Starcoder
 
-        .. autoclass:: bigdl.llm.langchain.embeddings.bigdlllm.StarcoderEmbeddings
+        .. autoclass:: ipex_llm.langchain.embeddings.bigdlllm.StarcoderEmbeddings
             :members:
             :undoc-members:
             :show-inheritance:

docs/readthedocs/source/doc/PythonAPI/LLM/optimize.rst

@@ -6,7 +6,7 @@ Optimize Model
 
 You can run any PyTorch model with ``optimize_model`` through only one-line code change to benefit from BigDL-LLM optimization, regardless of the library or API you are using.
 
-.. automodule:: bigdl.llm
+.. automodule:: ipex_llm
     :members: optimize_model
     :undoc-members:
     :show-inheritance:
@@ -18,7 +18,7 @@ Load Optimized Model
 
 To avoid high resource consumption during the loading processes of the original model, we provide save/load API to support the saving of model after low-bit optimization and the loading of the saved low-bit model. Saving and loading operations are platform-independent, regardless of their operating systems.
 
-.. automodule:: bigdl.llm.optimize
+.. automodule:: ipex_llm.optimize
     :members: load_low_bit
     :undoc-members:
     :show-inheritance:

docs/readthedocs/source/doc/PythonAPI/LLM/transformers.rst

@@ -10,7 +10,7 @@ You can apply BigDL-LLM optimizations on any Hugging Face Transformers models by
 AutoModelForCausalLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: bigdl.llm.transformers.AutoModelForCausalLM
+.. autoclass:: ipex_llm.transformers.AutoModelForCausalLM
     :members:
     :undoc-members:
     :show-inheritance:
@@ -22,7 +22,7 @@ AutoModelForCausalLM
 AutoModel
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: bigdl.llm.transformers.AutoModel
+.. autoclass:: ipex_llm.transformers.AutoModel
     :members:
     :undoc-members:
     :show-inheritance:
@@ -34,7 +34,7 @@ AutoModel
 AutoModelForSpeechSeq2Seq
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: bigdl.llm.transformers.AutoModelForSpeechSeq2Seq
+.. autoclass:: ipex_llm.transformers.AutoModelForSpeechSeq2Seq
     :members:
     :undoc-members:
     :show-inheritance:
@@ -46,7 +46,7 @@ AutoModelForSpeechSeq2Seq
 AutoModelForSeq2SeqLM
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. autoclass:: bigdl.llm.transformers.AutoModelForSeq2SeqLM
+.. autoclass:: ipex_llm.transformers.AutoModelForSeq2SeqLM
     :members:
     :undoc-members:
     :show-inheritance:
@@ -67,7 +67,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Llama
 
-        .. autoclass:: bigdl.llm.transformers.LlamaForCausalLM
+        .. autoclass:: ipex_llm.transformers.LlamaForCausalLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -77,7 +77,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: ChatGLM
 
-        .. autoclass:: bigdl.llm.transformers.ChatGLMForCausalLM
+        .. autoclass:: ipex_llm.transformers.ChatGLMForCausalLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -87,7 +87,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Gptneox
 
-        .. autoclass:: bigdl.llm.transformers.GptneoxForCausalLM
+        .. autoclass:: ipex_llm.transformers.GptneoxForCausalLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -96,7 +96,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
             .. automethod:: from_pretrained
 
     .. tab:: Bloom
-        .. autoclass:: bigdl.llm.transformers.BloomForCausalLM
+        .. autoclass:: ipex_llm.transformers.BloomForCausalLM
             :members:
             :undoc-members:
             :show-inheritance:
@@ -106,7 +106,7 @@ For ``llama``/``chatglm``/``bloom``/``gptneox``/``starcoder`` model families, yo
 
     .. tab:: Starcoder
 
-        .. autoclass:: bigdl.llm.transformers.StarcoderForCausalLM
+        .. autoclass:: ipex_llm.transformers.StarcoderForCausalLM
             :members:
             :undoc-members:
             :show-inheritance:

docs/readthedocs/source/index.rst

@@ -112,7 +112,7 @@ You can then apply INT4 optimizations to any Hugging Face *Transformers* models
 .. code-block:: python
 
    #load Hugging Face Transformers model with INT4 optimizations
-   from bigdl.llm.transformers import AutoModelForCausalLM
+   from ipex_llm.transformers import AutoModelForCausalLM
    model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
    #run the optimized model on Intel CPU
@@ -146,7 +146,7 @@ You can then apply INT4 optimizations to any Hugging Face *Transformers* models
 .. code-block:: python
 
    #load Hugging Face Transformers model with INT4 optimizations
-   from bigdl.llm.transformers import AutoModelForCausalLM
+   from ipex_llm.transformers import AutoModelForCausalLM
    model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
    #run the optimized model on Intel GPU

python/llm/README.md

@@ -146,7 +146,7 @@ You may apply INT4 optimizations to any Hugging Face *Transformers* model on Int
 
 ```python
 #load Hugging Face Transformers model with INT4 optimizations
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
 #run the optimized model on Intel CPU
@@ -164,7 +164,7 @@ You may apply INT4 optimizations to any Hugging Face *Transformers* model on Int
 
 ```python
 #load Hugging Face Transformers model with INT4 optimizations
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 import intel_extension_for_pytorch
 model = AutoModelForCausalLM.from_pretrained('/path/to/model/', load_in_4bit=True)
 
@@ -206,13 +206,13 @@ You may also convert Hugging Face *Transformers* models into native INT4 model f
   
 ```python
 #convert the model
-from bigdl.llm import llm_convert
+from ipex_llm import llm_convert
 bigdl_llm_path = llm_convert(model='/path/to/model/',
         outfile='/path/to/output/', outtype='int4', model_family="llama")
 
 #load the converted model
 #switch to ChatGLMForCausalLM/GptneoxForCausalLM/BloomForCausalLM/StarcoderForCausalLM to load other models
-from bigdl.llm.transformers import LlamaForCausalLM
+from ipex_llm.transformers import LlamaForCausalLM
 llm = LlamaForCausalLM.from_pretrained("/path/to/output/model.bin", native=True, ...)
   
 #run the converted model
@@ -231,8 +231,8 @@ You may run the models using the LangChain API in `bigdl-llm`.
   You may run any Hugging Face *Transformers* model (with INT4 optimiztions applied) using the LangChain API as follows:
 
   ```python
-  from bigdl.llm.langchain.llms import TransformersLLM
+  from ipex_llm.langchain.llms import TransformersLLM
-  from bigdl.llm.langchain.embeddings import TransformersEmbeddings
+  from ipex_llm.langchain.embeddings import TransformersEmbeddings
   from langchain.chains.question_answering import load_qa_chain
 
   embeddings = TransformersEmbeddings.from_model_id(model_id=model_path)
@@ -250,8 +250,8 @@ You may run the models using the LangChain API in `bigdl-llm`.
   >**Notes**:* Currently only llama/bloom/gptneox/starcoder/chatglm model families are supported; for other models, you may use the Hugging Face `transformers` model format as described above).
 
   ```python
-  from bigdl.llm.langchain.llms import LlamaLLM
+  from ipex_llm.langchain.llms import LlamaLLM
-  from bigdl.llm.langchain.embeddings import LlamaEmbeddings
+  from ipex_llm.langchain.embeddings import LlamaEmbeddings
   from langchain.chains.question_answering import load_qa_chain
 
   #switch to ChatGLMEmbeddings/GptneoxEmbeddings/BloomEmbeddings/StarcoderEmbeddings to load other models

python/llm/dev/benchmark/README.md

@@ -7,7 +7,7 @@ Just put this file into your benchmark directory, and then wrap your transformer
 Take `chatglm-6b` as an example:
 ```python
 import torch
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 from benchmark_util import BenchmarkWrapper
 
@@ -35,7 +35,7 @@ Take `chatglm-6b` as an example:
 ```python
 import torch
 import intel_extension_for_pytorch as ipex
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 from benchmark_util import BenchmarkWrapper
 

python/llm/dev/benchmark/all-in-one/run-stress-test.py

@@ -31,7 +31,7 @@ benchmark_util_path = os.path.join(current_dir, '..')
 import sys
 sys.path.append(benchmark_util_path)
 from benchmark_util import BenchmarkWrapper
-from bigdl.llm.utils.common.log4Error import invalidInputError
+from ipex_llm.utils.common.log4Error import invalidInputError
 
 LLAMA_IDS = ['meta-llama/Llama-2-7b-chat-hf','meta-llama/Llama-2-13b-chat-hf',
              'meta-llama/Llama-2-70b-chat-hf','decapoda-research/llama-7b-hf',
@@ -85,7 +85,7 @@ def run_transformer_int4(repo_id,
                          num_trials,
                          num_beams,
                          low_bit):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     model_path = get_model_path(repo_id, local_model_hub)
@@ -149,7 +149,7 @@ def run_transformer_int4_gpu(repo_id,
                              num_trials,
                              num_beams,
                              low_bit):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer
     import intel_extension_for_pytorch as ipex
     reserved_mem_list = []

python/llm/dev/benchmark/all-in-one/run.py

@@ -32,7 +32,7 @@ benchmark_util_path = os.path.join(current_dir, '..')
 import sys
 sys.path.append(benchmark_util_path)
 from benchmark_util import BenchmarkWrapper
-from bigdl.llm.utils.common.log4Error import invalidInputError
+from ipex_llm.utils.common.log4Error import invalidInputError
 
 LLAMA_IDS = ['meta-llama/Llama-2-7b-chat-hf','meta-llama/Llama-2-13b-chat-hf',
              'meta-llama/Llama-2-70b-chat-hf','decapoda-research/llama-7b-hf',
@@ -143,8 +143,8 @@ def run_native_int4(repo_id,
                     warm_up,
                     num_trials):
     model_path = get_model_path(repo_id, local_model_hub)
-    from bigdl.llm.transformers import BigdlNativeForCausalLM
+    from ipex_llm.transformers import BigdlNativeForCausalLM
-    from bigdl.llm import llm_convert
+    from ipex_llm import llm_convert
     if "chatglm" in repo_id.lower():
         family = "chatglm"
     elif "llama" in repo_id.lower():
@@ -184,7 +184,7 @@ def run_transformer_int4(repo_id,
                          num_beams,
                          low_bit,
                          batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     model_path = get_model_path(repo_id, local_model_hub)
@@ -319,7 +319,7 @@ def run_optimize_model(repo_id,
                        low_bit,
                        batch_size):
     from transformers import AutoModel, AutoModelForCausalLM, AutoTokenizer, LlamaTokenizer
-    from bigdl.llm import optimize_model
+    from ipex_llm import optimize_model
 
     model_path = get_model_path(repo_id, local_model_hub)
     # Load model in 4 bit,
@@ -389,7 +389,7 @@ def run_transformer_int4_gpu(repo_id,
                              num_beams,
                              low_bit,
                              batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
@@ -495,7 +495,7 @@ def run_optimize_model_gpu(repo_id,
                            low_bit,
                            batch_size):
     from transformers import AutoModel, AutoModelForCausalLM, AutoTokenizer, GPTJForCausalLM, LlamaTokenizer
-    from bigdl.llm import optimize_model
+    from ipex_llm import optimize_model
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
     # Load model in 4 bit,
@@ -651,7 +651,7 @@ def run_bigdl_fp16_gpu(repo_id,
                        num_trials,
                        num_beams,
                        batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
@@ -731,7 +731,7 @@ def run_deepspeed_transformer_int4_cpu(repo_id,
                          batch_size):
     from transformers import AutoModelForCausalLM, LlamaTokenizer, AutoTokenizer
     import deepspeed
-    from bigdl.llm import optimize_model
+    from ipex_llm import optimize_model
     import argparse
     # parser is for deepspeed subprocesses' inline parameter
     parser = argparse.ArgumentParser(description='Predict Tokens using `generate()` API for Llama2 model')
@@ -822,7 +822,7 @@ def run_transformer_int4_gpu_win(repo_id,
                                  cpu_embedding,
                                  batch_size,
                                  streaming):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer, TextStreamer
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
@@ -928,7 +928,7 @@ def run_transformer_int4_fp16_gpu_win(repo_id,
                                       cpu_embedding,
                                       batch_size,
                                       streaming):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer, TextStreamer
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
@@ -1038,7 +1038,7 @@ def run_transformer_int4_loadlowbit_gpu_win(repo_id,
                                             cpu_embedding,
                                             batch_size,
                                             streaming):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, GPTJForCausalLM, LlamaTokenizer, TextStreamer
     import intel_extension_for_pytorch as ipex
     model_path = get_model_path(repo_id, local_model_hub)
@@ -1140,7 +1140,7 @@ def run_transformer_autocast_bf16( repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     model_path = get_model_path(repo_id, local_model_hub)
@@ -1209,7 +1209,7 @@ def run_bigdl_ipex_bf16(repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     os.environ["BIGDL_OPT_IPEX"] = "true"
@@ -1280,7 +1280,7 @@ def run_bigdl_ipex_int4(repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     os.environ["BIGDL_OPT_IPEX"] = "true"
@@ -1350,7 +1350,7 @@ def run_bigdl_ipex_int8(repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     os.environ["BIGDL_OPT_IPEX"] = "true"
@@ -1434,7 +1434,7 @@ def run_deepspeed_optimize_model_gpu(repo_id,
     os.environ["MASTER_PORT"] = os.environ.get("MASTER_PORT", "29500")
 
     from transformers import AutoModel, AutoModelForCausalLM, AutoTokenizer, GPTJForCausalLM, LlamaTokenizer
-    from bigdl.llm import optimize_model
+    from ipex_llm import optimize_model
     import intel_extension_for_pytorch as ipex
     import deepspeed
     from deepspeed.accelerator.cpu_accelerator import CPU_Accelerator
@@ -1535,9 +1535,9 @@ def run_speculative_cpu(repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
-    from bigdl.llm.transformers.convert import get_enable_ipex
+    from ipex_llm.transformers.convert import get_enable_ipex
 
     _enable_ipex = get_enable_ipex()
 
@@ -1615,7 +1615,7 @@ def run_speculative_gpu(repo_id,
                     num_trials,
                     num_beams,
                     batch_size):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
 
     model_path = get_model_path(repo_id, local_model_hub)

python/llm/dev/benchmark/all-in-one/save.py

@@ -30,7 +30,7 @@ current_dir = os.path.dirname(os.path.realpath(__file__))
 def save_model_in_low_bit(repo_id,
                           local_model_hub,
                           low_bit):
-    from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
     from transformers import AutoTokenizer, LlamaTokenizer
     model_path = get_model_path(repo_id, local_model_hub)
     # Load model in 4 bit,

python/llm/dev/benchmark/ceval/eval.py

@@ -21,7 +21,7 @@ import torch
 import json
 from tqdm import tqdm
 
-from bigdl.llm.utils.common.log4Error import invalidInputError
+from ipex_llm.utils.common.log4Error import invalidInputError
 from evaluators.qwen import QwenEvaluator
 from evaluators.llama import LlamaEvaluator
 from evaluators.chatglm import ChatGLMEvaluator

python/llm/dev/benchmark/ceval/evaluators/chatglm.py

@@ -22,7 +22,7 @@ from thefuzz import process
 from transformers import AutoTokenizer
 
 from evaluators.evaluator import Evaluator
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers.generation.utils import LogitsProcessorList
 from transformers.generation.logits_process import LogitsProcessor
 

python/llm/dev/benchmark/ceval/evaluators/llama.py

@@ -22,7 +22,7 @@ import numpy as np
 import torch
 from transformers import LlamaTokenizer, GenerationConfig
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from evaluators.evaluator import Evaluator
 
 

python/llm/dev/benchmark/ceval/evaluators/qwen.py

@@ -22,7 +22,7 @@ from thefuzz import process
 from transformers import AutoTokenizer
 from transformers.generation import GenerationConfig
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from evaluators.evaluator import Evaluator
 
 

python/llm/dev/benchmark/harness/bigdl_llm.py

@@ -14,7 +14,7 @@
 # limitations under the License.
 #
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 import inspect
 from lm_eval.models.huggingface import AutoCausalLM

python/llm/dev/benchmark/perplexity/ppl.py

@@ -20,7 +20,7 @@ from torch.nn import CrossEntropyLoss
 from tqdm import tqdm
 import gc
 
-from bigdl.llm.transformers import AutoModelForCausalLM, AutoModel
+from ipex_llm.transformers import AutoModelForCausalLM, AutoModel
 
 class BigDLPPL:
     def __init__(self, model_path, device, **model_kwargs) -> None:

python/llm/dev/benchmark/perplexity/run.py

@@ -21,7 +21,7 @@ from datasets import concatenate_datasets, load_dataset
 from transformers import AutoTokenizer
 
 from ppl import BigDLPPL
-from bigdl.llm.ggml.quantize import ggml_tensor_qtype
+from ipex_llm.ggml.quantize import ggml_tensor_qtype
 
 import os
 import json

python/llm/dev/benchmark/whisper/run_whisper.py

@@ -15,7 +15,7 @@
 #
 
 from datasets import load_dataset
-from bigdl.llm.transformers import AutoModelForSpeechSeq2Seq
+from ipex_llm.transformers import AutoModelForSpeechSeq2Seq
 from transformers import WhisperProcessor
 import torch
 from evaluate import load

python/llm/example/CPU/Applications/autogen/README.md

@@ -69,11 +69,11 @@ conda activate autogen
 cd autogen
 
 # load the local model with cpu with your downloaded model
-python -m bigdl.llm.serving.model_worker --model-path ... --device cpu
+python -m ipex_llm.serving.model_worker --model-path ... --device cpu
 ```
 
 Change the Model Name:
-> Assume you use the model `Mistral-7B-Instruct-v0.2` and your model is downloaded to `autogen/model/Mistral-7B-Instruct-v0.2`. You should rename the model to `autogen/model/bigdl` and run `python -m bigdl.llm.serving.model_worker --model-path ... --device cpu`. This ensures the proper usage of the BigDL-adapted FastChat.
+> Assume you use the model `Mistral-7B-Instruct-v0.2` and your model is downloaded to `autogen/model/Mistral-7B-Instruct-v0.2`. You should rename the model to `autogen/model/bigdl` and run `python -m ipex_llm.serving.model_worker --model-path ... --device cpu`. This ensures the proper usage of the BigDL-adapted FastChat.
 
 Potential Error Note:
 > If you get `RuntimeError: Error register to Controller` in the worker terminal, please set `export no_proxy='localhost'` to ensure the registration

python/llm/example/CPU/Applications/hf-agent/run_agent.py

@@ -19,7 +19,7 @@ import argparse
 from PIL import Image
 from transformers import AutoTokenizer, LocalAgent
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 if __name__ == "__main__":
     parser = argparse.ArgumentParser(description="Run agent using vicuna model")

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

@@ -3,7 +3,7 @@
 In this example, we apply low-bit optimizations to [Streaming-LLM](https://github.com/mit-han-lab/streaming-llm/tree/main#efficient-streaming-language-models-with-attention-sinks) using BigDL-LLM, which can deploy low-bit(including FP4/INT4/FP8/INT8) LLMs for infinite-length inputs.
 Only one code change is needed to load the model using bigdl-llm as follows:
 ```python
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 model = AutoModelForCausalLM.from_pretrained(model_name_or_path, load_in_4bit=True, trust_remote_code=True, optimize_model=False)
 ```
 

python/llm/example/CPU/Applications/streaming-llm/streaming_llm/utils.py

@@ -49,7 +49,7 @@ import urllib.request
 import os
 import json
 # code change to import from bigdl-llm API instead of using transformers API
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer
 import intel_extension_for_pytorch as ipex
 

python/llm/example/CPU/Deepspeed-AutoTP/README.md

@@ -39,7 +39,7 @@ Distributed model managed by deepspeed can be further optimized with BigDL low-b
 
 ```python
 # Apply BigDL-LLM INT4 optimizations on transformers
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 
 model = optimize_model(model.module.to(f'cpu'), low_bit='sym_int4')
 model = model.to(f'cpu:{local_rank}') # move partial model to local rank

python/llm/example/CPU/Deepspeed-AutoTP/deepspeed_autotp.py

@@ -45,7 +45,7 @@ import os
 import torch
 from transformers import AutoModelForCausalLM, LlamaTokenizer, AutoTokenizer
 import deepspeed
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 import torch
 import intel_extension_for_pytorch as ipex
 import time

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/AWQ/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/GGUF/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 # you could tune the prompt based on your own model,
 # here the prompt tuning refers to https://huggingface.co/georgesung/llama2_7b_chat_uncensored#prompt-style

python/llm/example/CPU/HF-Transformers-AutoModels/Advanced-Quantizations/GPTQ/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer, GPTQConfig
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/aquila/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/aquila2/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/baichuan/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/baichuan2/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # prompt format referred from https://github.com/baichuan-inc/Baichuan2/issues/227 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/bluelm/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm2/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm2/streamchat.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm3/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/chatglm3/streamchat.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel
+from ipex_llm.transformers import AutoModel
 from transformers import AutoTokenizer
 
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/codellama/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import CodeLlamaTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/codeshell/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deciLM-7b/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 
 from transformers import AutoTokenizer
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 # you could tune the prompt based on your own model,
 # here the prompt tuning refers to https://huggingface.co/Deci/DeciLM-7B-instruct#prompt-template

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deepseek-moe/generate.py

@@ -36,7 +36,7 @@ if __name__ == '__main__':
 
     # Load model in 4 bit,
     # which convert the relevant layers in the model into INT4 format
-    from bigdl.llm.transformers import AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModelForCausalLM
     model = AutoModelForCausalLM.from_pretrained(model_path, trust_remote_code=True, load_in_4bit=True).eval()
     model.generation_config = GenerationConfig.from_pretrained(model_path)
     model.generation_config.pad_token_id = model.generation_config.eos_token_id

python/llm/example/CPU/HF-Transformers-AutoModels/Model/deepseek/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/distil-whisper/recognize.py

@@ -17,7 +17,7 @@
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForSpeechSeq2Seq
+from ipex_llm.transformers import AutoModelForSpeechSeq2Seq
 from datasets import load_dataset
 from transformers import pipeline
 from transformers.models.whisper import WhisperFeatureExtractor, WhisperTokenizer

python/llm/example/CPU/HF-Transformers-AutoModels/Model/dolly_v1/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/dolly_v2/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/falcon/falcon-7b-instruct/modelling_RW.py

@@ -301,7 +301,7 @@ class Attention(nn.Module):
             # resize qk to 4D to match apply_rotary_pos_emb_no_cache_xpu's requirements.
             query_layer = query_layer.reshape(batch_size, self.num_heads, q_length, self.head_dim)
             key_layer = key_layer.reshape(batch_size, self.num_kv, q_length, self.head_dim)
-            from bigdl.llm.transformers.models.utils import apply_rotary_pos_emb_no_cache_xpu
+            from ipex_llm.transformers.models.utils import apply_rotary_pos_emb_no_cache_xpu
             query_layer, key_layer = apply_rotary_pos_emb_no_cache_xpu(query_layer,
                                                                        key_layer,
                                                                        position_ids,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/falcon/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/flan-t5/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForSeq2SeqLM
+from ipex_llm.transformers import AutoModelForSeq2SeqLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/fuyu/generate.py

@@ -19,7 +19,7 @@ import torch
 import argparse
 import time
 from PIL import Image
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser(description='Predict Tokens using `generate()` API for Fuyu model')

python/llm/example/CPU/HF-Transformers-AutoModels/Model/gemma/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # The instruction-tuned models use a chat template that must be adhered to for conversational use.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm-xcomposer/chat.py

@@ -14,14 +14,14 @@
 # limitations under the License.
 #
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 from transformers.generation import GenerationConfig
 import torch
 import time
 import os
 import argparse
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser(description='Predict Tokens using `chat()` API for InternLM-XComposer model')

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/internlm2/generate.py

@@ -39,7 +39,7 @@ if __name__ == '__main__':
     model_path = args.repo_id_or_model_path
 
     
-    from bigdl.llm.transformers import AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModelForCausalLM
     model = AutoModelForCausalLM.from_pretrained(model_path,
                                                  load_in_4bit=True,
                                                  trust_remote_code=True)

python/llm/example/CPU/HF-Transformers-AutoModels/Model/llama2/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mistral/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mixtral/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/moss/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/mpt/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer, GenerationConfig
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phi-1_5/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel,AutoModelForCausalLM
+from ipex_llm.transformers import AutoModel,AutoModelForCausalLM
 from transformers import AutoTokenizer, GenerationConfig
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phi-2/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModel,AutoModelForCausalLM
+from ipex_llm.transformers import AutoModel,AutoModelForCausalLM
 from transformers import AutoTokenizer, GenerationConfig
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phixtral/generate.py

@@ -20,7 +20,7 @@ import argparse
 import numpy as np
 
 from transformers import AutoTokenizer, GenerationConfig
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 # you could tune the prompt based on your own model,
 # here the prompt tuning refers to  # TODO: https://huggingface.co/microsoft/phi-1_5/blob/main/modeling_mixformer_sequential.py
 PHI1_5_PROMPT_FORMAT = " Question:{prompt}\n\n Answer:"
@@ -41,7 +41,7 @@ if __name__ == '__main__':
 
     # Load model in 4 bit,
     # which convert the relevant layers in the model into INT4 format
-    from bigdl.llm.transformers import AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModelForCausalLM
     model = AutoModelForCausalLM.from_pretrained(model_path,
                                                  load_in_4bit=True,
                                                  trust_remote_code=True)

python/llm/example/CPU/HF-Transformers-AutoModels/Model/phoenix/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could revise it based on the Phoenix model you choose to use

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen-vl/chat.py

@@ -14,14 +14,14 @@
 # limitations under the License.
 #
 
-from bigdl.llm.transformers import AutoModel, AutoModelForCausalLM
+from ipex_llm.transformers import AutoModel, AutoModelForCausalLM
 from transformers import AutoTokenizer, LlamaTokenizer
 from transformers.generation import GenerationConfig
 import torch
 import time
 import os
 import argparse
-from bigdl.llm import optimize_model
+from ipex_llm import optimize_model
 torch.manual_seed(1234)
 
 if __name__ == '__main__':

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model

python/llm/example/CPU/HF-Transformers-AutoModels/Model/qwen1.5/generate.py

@@ -36,7 +36,7 @@ if __name__ == '__main__':
     model_path = args.repo_id_or_model_path
 
     
-    from bigdl.llm.transformers import AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModelForCausalLM
     model = AutoModelForCausalLM.from_pretrained(model_path,
                                                  load_in_4bit=True,
                                                  trust_remote_code=True)

python/llm/example/CPU/HF-Transformers-AutoModels/Model/redpajama/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/replit/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/skywork/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/solar/generate.py

@@ -19,7 +19,7 @@ import time
 import argparse
 import numpy as np
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/starcoder/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/vicuna/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer
 
 # you could tune the prompt based on your own model,

python/llm/example/CPU/HF-Transformers-AutoModels/Model/whisper/long-segment-recognize.py

@@ -19,7 +19,7 @@ import librosa
 import argparse
 
 from transformers import pipeline
-from bigdl.llm.transformers import AutoModelForSpeechSeq2Seq
+from ipex_llm.transformers import AutoModelForSpeechSeq2Seq
 from transformers.models.whisper import WhisperFeatureExtractor, WhisperTokenizer
 
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/whisper/recognize.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForSpeechSeq2Seq
+from ipex_llm.transformers import AutoModelForSpeechSeq2Seq
 from transformers import WhisperProcessor
 from datasets import load_dataset
 

python/llm/example/CPU/HF-Transformers-AutoModels/Model/wizardcoder-python/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer
 
 WIZARDCODERPYTHON_PROMPT_FORMAT = """Below is an instruction that describes a task. Write a response that appropriately completes the request.

python/llm/example/CPU/HF-Transformers-AutoModels/Model/yi/generate.py

@@ -18,7 +18,7 @@ import torch
 import time
 import argparse
 
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import AutoTokenizer
 
 # Refer to https://huggingface.co/01-ai/Yi-6B-Chat#31-use-the-chat-model

python/llm/example/CPU/HF-Transformers-AutoModels/Model/yuan2/generate.py

@@ -18,7 +18,7 @@ import torch, transformers
 import sys, os, time
 import argparse
 from transformers import LlamaTokenizer
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 
 # Refer to https://huggingface.co/IEITYuan/Yuan2-2B-hf#Usage
 YUAN2_PROMPT_FORMAT = """

python/llm/example/CPU/HF-Transformers-AutoModels/Model/ziya/generate.py

@@ -39,7 +39,7 @@ if __name__ == '__main__':
     model_path = args.repo_id_or_model_path
 
     
-    from bigdl.llm.transformers import AutoModelForCausalLM
+    from ipex_llm.transformers import AutoModelForCausalLM
     # enabling `use_cache=True` allows the model to utilize the previous
     # key/values attentions to speed up decoding;
     # to obtain optimal performance with BigDL-LLM INT4 optimizations,

python/llm/example/CPU/HF-Transformers-AutoModels/More-Data-Types/transformers_low_bit_pipeline.py

@@ -15,7 +15,7 @@
 #
 
 import argparse
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer, TextGenerationPipeline
 
 if __name__ == '__main__':
@@ -38,7 +38,7 @@ if __name__ == '__main__':
         model = AutoModelForCausalLM.load_low_bit(load_path)
         tokenizer = LlamaTokenizer.from_pretrained(load_path)
     else:
-        # load_in_low_bit in bigdl.llm.transformers will convert
+        # load_in_low_bit in ipex_llm.transformers will convert
         # the relevant layers in the model into corresponding int X format
         model = AutoModelForCausalLM.from_pretrained(model_path, load_in_low_bit=low_bit, trust_remote_code=True)
         tokenizer = LlamaTokenizer.from_pretrained(model_path, trust_remote_code=True)

python/llm/example/CPU/HF-Transformers-AutoModels/Save-Load/transformers_low_bit_pipeline.py

@@ -15,7 +15,7 @@
 #
 
 import argparse
-from bigdl.llm.transformers import AutoModelForCausalLM
+from ipex_llm.transformers import AutoModelForCausalLM
 from transformers import LlamaTokenizer, TextGenerationPipeline
 
 if __name__ == '__main__':
@@ -38,7 +38,7 @@ if __name__ == '__main__':
         model = AutoModelForCausalLM.load_low_bit(load_path)
         tokenizer = LlamaTokenizer.from_pretrained(load_path)
     else:
-        # load_in_low_bit in bigdl.llm.transformers will convert
+        # load_in_low_bit in ipex_llm.transformers will convert
         # the relevant layers in the model into corresponding int X format
         model = AutoModelForCausalLM.from_pretrained(model_path, load_in_low_bit=low_bit, trust_remote_code=True)
         tokenizer = LlamaTokenizer.from_pretrained(model_path, trust_remote_code=True)

python/llm/example/CPU/LangChain/native_int4/docqa.py

@@ -31,8 +31,8 @@ from langchain.chains.question_answering import load_qa_chain
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 
-from bigdl.llm.langchain.llms import *
+from ipex_llm.langchain.llms import *
-from bigdl.llm.langchain.embeddings import *
+from ipex_llm.langchain.embeddings import *
 
 
 def main(args):

python/llm/example/CPU/LangChain/native_int4/streamchat.py

@@ -21,7 +21,7 @@
 
 import argparse
 
-from bigdl.llm.langchain.llms import *
+from ipex_llm.langchain.llms import *
 from langchain import PromptTemplate, LLMChain
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler

python/llm/example/CPU/LangChain/native_int4/voiceassistant.py

@@ -23,7 +23,7 @@
 
 
 from langchain import LLMChain, PromptTemplate
-from bigdl.llm.langchain.llms import *
+from ipex_llm.langchain.llms import *
 from langchain.memory import ConversationBufferWindowMemory
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler

python/llm/example/CPU/LangChain/transformers_int4/chat.py

@@ -21,7 +21,7 @@
 
 import argparse
 
-from bigdl.llm.langchain.llms import TransformersLLM, TransformersPipelineLLM
+from ipex_llm.langchain.llms import TransformersLLM, TransformersPipelineLLM
 from langchain import PromptTemplate, LLMChain
 from langchain import HuggingFacePipeline
 

python/llm/example/CPU/LangChain/transformers_int4/llm_math.py

@@ -25,7 +25,7 @@
 import argparse
 
 from langchain.chains import LLMMathChain
-from bigdl.llm.langchain.llms import TransformersLLM, TransformersPipelineLLM
+from ipex_llm.langchain.llms import TransformersLLM, TransformersPipelineLLM
 
 
 def main(args):

python/llm/example/CPU/LangChain/transformers_int4/rag.py

@@ -30,8 +30,8 @@ from langchain.text_splitter import CharacterTextSplitter
 from langchain.chains.question_answering import load_qa_chain
 from langchain.callbacks.manager import CallbackManager
 
-from bigdl.llm.langchain.llms import TransformersLLM
+from ipex_llm.langchain.llms import TransformersLLM
-from bigdl.llm.langchain.embeddings import TransformersEmbeddings
+from ipex_llm.langchain.embeddings import TransformersEmbeddings
 
 text_doc = '''
 BigDL seamlessly scales your data analytics & AI applications from laptop to cloud, with the following libraries:

python/llm/example/CPU/LangChain/transformers_int4/voiceassistant.py

@@ -23,9 +23,9 @@
 
 
 from langchain import LLMChain, PromptTemplate
-from bigdl.llm.langchain.llms import TransformersLLM
+from ipex_llm.langchain.llms import TransformersLLM
 from langchain.memory import ConversationBufferWindowMemory
-from bigdl.llm.transformers import AutoModelForSpeechSeq2Seq
+from ipex_llm.transformers import AutoModelForSpeechSeq2Seq
 from transformers import WhisperProcessor
 import speech_recognition as sr
 import numpy as np