cohere

CoHere/command-r

In this directory, you will find examples on how you could apply IPEX-LLM INT4 optimizations on cohere models on Intel GPUs. For illustration purposes, we utilize the CohereForAI/c4ai-command-r-v01 as a reference model.

Note: Because the size of this cohere model is 35B, even running low_bit sym_int4 still requires about 17.5GB. So currently it can only be run on MAX GPU, or run with Pipeline-Parallel-Inference on multiple Arc GPUs.

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

0. Requirements

To run these examples with IPEX-LLM on Intel GPUs, 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 cohere model to predict the next N tokens using generate() API, with IPEX-LLM INT4 optimizations on Intel GPUs.

1. Install

1.1 Installation on Linux

We suggest using conda to manage environment:

conda create -n llm python=3.11
conda activate llm
# below command will install intel_extension_for_pytorch==2.1.10+xpu as default
pip install --pre --upgrade ipex-llm[xpu] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
pip install transformers==4.40.0
conda install -c conda-forge -y gperftools=2.10 # to enable tcmalloc

1.2 Installation on Windows

We suggest using conda to manage environment:

conda create -n llm python=3.11 libuv
conda activate llm

# below command will install intel_extension_for_pytorch==2.1.10+xpu as default
pip install --pre --upgrade ipex-llm[xpu] --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/xpu/us/
pip install transformers==4.40.0

2. Configures OneAPI environment variables for Linux

Note

Skip this step if you are running on Windows.

This is a required step on Linux for APT or offline installed oneAPI. Skip this step for PIP-installed oneAPI.

source /opt/intel/oneapi/setvars.sh

3. Runtime Configurations

For optimal performance, it is recommended to set several environment variables. Please check out the suggestions based on your device.

3.1 Configurations for Linux

For Intel Arc™ A-Series Graphics and Intel Data Center GPU Flex Series

export USE_XETLA=OFF
export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
export SYCL_CACHE_PERSISTENT=1

For Intel Data Center GPU Max Series

export LD_PRELOAD=${LD_PRELOAD}:${CONDA_PREFIX}/lib/libtcmalloc.so
export SYCL_PI_LEVEL_ZERO_USE_IMMEDIATE_COMMANDLISTS=1
export SYCL_CACHE_PERSISTENT=1
export ENABLE_SDP_FUSION=1

Note: Please note that libtcmalloc.so can be installed by conda install -c conda-forge -y gperftools=2.10.

4. Running examples

python ./generate.py --repo-id-or-model-path REPO_ID_OR_MODEL_PATH --prompt PROMPT --n-predict N_PREDICT

Arguments info:

• --repo-id-or-model-path REPO_ID_OR_MODEL_PATH: argument defining the huggingface repo id for the cohere model to be downloaded, or the path to the huggingface checkpoint folder. It is default to be 'CohereForAI/c4ai-command-r-v01'.
• --prompt PROMPT: argument defining the prompt to be infered (with integrated prompt format for chat). It is default to be 'What is AI?'.
• --n-predict N_PREDICT: argument defining the max number of tokens to predict. It is default to be 32.

Sample Output

CohereForAI/c4ai-command-r-v01

Inference time: xxxxx s
-------------------- Prompt --------------------

<BOS_TOKEN><|START_OF_TURN_TOKEN|><|USER_TOKEN|>What is AI?<|END_OF_TURN_TOKEN|><|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>

-------------------- Output --------------------

<|START_OF_TURN_TOKEN|><|USER_TOKEN|>What is AI?<|START_OF_TURN_TOKEN|><|CHATBOT_TOKEN|>
Artificial Intelligence Quora User,

Artificial Intelligence (AI) is the simulation of human intelligence in machines, typically by machines, that have become a very complex