Accelerate documentation

Installation and Configuration

Hugging Face's logo
Join the Hugging Face community

and get access to the augmented documentation experience

to get started

Installation and Configuration

Before you start, you will need to setup your environment, install the appropriate packages, and configure 🤗 Accelerate. 🤗 Accelerate is tested on Python 3.8+.

Installing 🤗 Accelerate

🤗 Accelerate is available on pypi and conda, as well as on GitHub. Details to install from each are below:


To install 🤗 Accelerate from pypi, perform:

pip install accelerate


🤗 Accelerate can also be installed with conda with:

conda install -c conda-forge accelerate


New features are added every day that haven’t been released yet. To try them out yourself, install from the GitHub repository:

pip install git+

If you’re working on contributing to the library or wish to play with the source code and see live results as you run the code, an editable version can be installed from a locally-cloned version of the repository:

git clone
cd accelerate
pip install -e .

Configuring 🤗 Accelerate

After installing, you need to configure 🤗 Accelerate for how the current system is setup for training. To do so run the following and answer the questions prompted to you:

accelerate config

To write a barebones configuration that doesn’t include options such as DeepSpeed configuration or running on TPUs, you can quickly run:

python -c "from accelerate.utils import write_basic_config; write_basic_config(mixed_precision='fp16')"

🤗 Accelerate will automatically utilize the maximum number of GPUs available and set the mixed precision mode.

To check that your configuration looks fine, run:

accelerate env

An example output is shown below, which describes two GPUs on a single machine with no mixed precision being used:

- `Accelerate` version: 0.11.0.dev0
- Platform: Linux-5.10.0-15-cloud-amd64-x86_64-with-debian-11.3
- Python version: 3.7.12
- Numpy version: 1.19.5
- PyTorch version (GPU?): 1.12.0+cu102 (True)
- `Accelerate` default config:
        - compute_environment: LOCAL_MACHINE
        - distributed_type: MULTI_GPU
        - mixed_precision: no
        - use_cpu: False
        - num_processes: 2
        - machine_rank: 0
        - num_machines: 1
        - main_process_ip: None
        - main_process_port: None
        - main_training_function: main
        - deepspeed_config: {}
        - fsdp_config: {}
< > Update on GitHub