chatglm

ChatGLM

In this directory, you will find examples on how you could apply IPEX-LLM INT4 optimizations on ChatGLM models. For illustration purposes, we utilize the THUDM/chatglm-6b as a reference ChatGLM model.

Note: If you want to download the Hugging Face Transformers model, please refer to here.

IPEX-LLM optimizes the Transformers model in INT4 precision at runtime, and thus no explicit conversion is needed.

Requirements

To run these examples with IPEX-LLM, we have some recommended requirements for your machine, please refer to here for more information.

Example: Predict Tokens using generate() API

In the example generate.py, we show a basic use case for a ChatGLM model to predict the next N tokens using generate() API, with IPEX-LLM INT4 optimizations.

1. Install

We suggest using conda to manage the Python environment. For more information about conda installation, please refer to here.

After installing conda, create a Python environment for IPEX-LLM:

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+

2. Run

After setting up the Python environment, you could run the example by following steps.

Note: When loading the model in 4-bit, IPEX-LLM converts linear layers in the model into INT4 format. In theory, a XB model saved in 16-bit will requires approximately 2X GB of memory for loading, and ~0.5X GB memory for further inference.

Please select the appropriate size of the ChatGLM model based on the capabilities of your machine.

2.1 Client

On client Windows machines, it is recommended to run directly with full utilization of all cores:

python ./generate.py --prompt 'AI是什么？'

More information about arguments can be found in Arguments Info section. The expected output can be found in Sample Output section.

2.2 Server

For optimal performance on server, it is recommended to set several environment variables (refer to here for more information), and run the example with all the physical cores of a single socket.

E.g. on Linux,

# set IPEX-LLM env variables
source ipex-llm-init

# e.g. for a server with 48 cores per socket
export OMP_NUM_THREADS=48
numactl -C 0-47 -m 0 python ./generate.py --prompt 'AI是什么？'

More information about arguments can be found in Arguments Info section. The expected output can be found in Sample Output section.

2.3 Arguments Info

In the example, several arguments can be passed to satisfy your requirements:

• --repo-id-or-model-path: str, argument defining the huggingface repo id for the ChatGLM model to be downloaded, or the path to the huggingface checkpoint folder. It is default to be 'THUDM/chatglm-6b'.
• --prompt: str, argument defining the prompt to be inferred (with integrated prompt format for chat). It is default to be 'AI是什么？'.
• --n-predict: int, argument defining the max number of tokens to predict. It is default to be 32.

2.4 Sample Output

THUDM/chatglm-6b

Inference time: xxxx s
-------------------- Output --------------------
问:AI是什么?
答: AI是人工智能(Artificial Intelligence)的缩写,指的是一种能够模拟人类智能的技术或系统。AI系统可以通过学习、推理、解决问题等方式,实现类似于