51 lines
1.2 KiB
ReStructuredText
51 lines
1.2 KiB
ReStructuredText
Installation
|
|
============
|
|
|
|
CacheFlow is a Python library that includes some C++ and CUDA code.
|
|
CacheFlow can run on systems that meet the following requirements:
|
|
|
|
* OS: Linux
|
|
* Python: 3.8 or higher
|
|
* CUDA: 11.0 -- 11.8
|
|
* GPU: compute capability 7.0 or higher (e.g., V100, T4, RTX20xx, A100, etc.)
|
|
|
|
.. note::
|
|
As of now, CacheFlow does not support CUDA 12.
|
|
If you are using Hopper or Lovelace GPUs, please use CUDA 11.8.
|
|
|
|
.. tip::
|
|
If you have trouble installing CacheFlow, we recommend using the NVIDIA PyTorch Docker image.
|
|
|
|
.. code-block:: console
|
|
|
|
$ docker run --gpus all -it --rm --shm-size=8g nvcr.io/nvidia/pytorch:22.12-py3
|
|
|
|
Install with pip
|
|
----------------
|
|
|
|
You can install CacheFlow using pip:
|
|
|
|
.. code-block:: console
|
|
|
|
$ # (Optional) Create a new conda environment.
|
|
$ conda create -n myenv python=3.8 -y
|
|
$ conda activate myenv
|
|
|
|
$ # Install CacheFlow.
|
|
$ pip install cacheflow
|
|
|
|
|
|
.. _build_from_source:
|
|
|
|
Build from source
|
|
-----------------
|
|
|
|
You can also build and install CacheFlow from source.
|
|
|
|
.. code-block:: console
|
|
|
|
$ git clone https://github.com/WoosukKwon/cacheflow.git
|
|
$ cd cacheflow
|
|
$ pip install -r requirements.txt
|
|
$ pip install -e . # This may take 5-10 minutes.
|