# Handling Spaces Dependencies in Gradio Spaces

## Default dependencies

The default Gradio Spaces environment comes with several pre-installed dependencies:

* The [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/index) client library allows you to manage your repository and files on the Hub with Python and programmatically access [Inference Providers](./models-inference) from your Space. If you choose to instantiate the model in your app with Inference Providers, you can benefit from the built-in acceleration optimizations. This option also consumes less computing resources, which is always nice for the environment! 🌎 

  Refer to this [page](https://huggingface.co/docs/huggingface_hub/how-to-inference) for more information on how to programmatically access Inference Providers.

* [`requests`](https://docs.python-requests.org/en/master/) is useful for calling third-party APIs from your app.

* [`datasets`](https://github.com/huggingface/datasets) allows you to fetch or display any dataset from the Hub inside your app.

* [`gradio`](https://github.com/gradio-app/gradio). You can optionally require a specific version using [`sdk_version` in the `README.md` file](spaces-config-reference).

* Common Debian packages, such as `ffmpeg`, `cmake`, `libsm6`, and few others.

## Adding your own dependencies

If you need other Python packages to run your app, add them to a **requirements.txt** file at the root of the repository. The Spaces runtime engine will create a custom environment on-the-fly. You can also add a **pre-requirements.txt** file describing dependencies that will be installed before your main dependencies. It can be useful if you need to update pip itself.

Debian dependencies are also supported. Add a **packages.txt** file at the root of your repository, and list all your dependencies in it. Each dependency should be on a separate line, and each line will be read and installed by `apt-get install`.