[Yandex Cloud documentation](../../../index.md) > [Yandex DataSphere](../../index.md) > [Concepts](../index.md) > [DataSphere Jobs](index.md) > Docker images in jobs

# Docker images in jobs

By default, DataSphere Jobs [jobs](index.md) use the `nvidia/cuda:11.8.0-cudnn8-runtime-ubuntu22.04` public image with a pre-installed Conda package manager, Python 3.10, and other additional packages. This image is stored in the DataSphere cache, so using the default environment allows you to run your jobs faster.

You can also use a different Docker image to run jobs by specifying it under `env` in the job configuration file. This can be:

* DataSphere system image

  ```text
  env:
    docker: system-python-3-10   # Python 3.10 system image
  ```

* Custom Docker image available in the job project

  ```text
  env:
    docker: <Docker_image_ID>  # ID expressed as bxxxxxxxxxxxxxxxxxxx
  ```

  {% note warning %}

  When using a project Docker image, the job runtime environment will not include libraries installed in the notebook.

  {% endnote %}

* External image

  You can use any preferred image registry ([Yandex Container Registry](https://yandex.cloud/en/services/container-registry), [Docker Hub](https://hub.docker.com/), [Docker — Private Registries](https://www.geeksforgeeks.org/docker-private-registries/), etc.) by specifying the username and password to access the image.

  ```text
  env:
    docker:
      image: <image_path>
      username: <login>
      password:
        secret-id: <secret_name>
  ```

  Where:

  * `<image_path>`: Full path to the image in a container registry, e.g., `cr.yandex/b1g**********/myenv:0.1`.
  * `<username>`: Username for access to your registry. For Yandex Container Registry authentication, use a [service account](../../../iam/concepts/users/service-accounts.md) and an [authorized key](../../../iam/concepts/authorization/key.md).
  * `<secret_name>`: Name of the DataSphere password secret. The secret must be [created](../../operations/data/secrets.md#create) in a DataSphere project.

  If you are using a public image, you do not need to specify authentication credentials:

  ```text
  env:
    docker:
      image: ubuntu:focal
  ```

#### Useful links {#see-also}

* [DataSphere Jobs](index.md)
* [DataSphere CLI](cli.md)
* [Job runtime environment](environment.md)
* [Running jobs in DataSphere Jobs](../../operations/projects/work-with-jobs.md)
* [GitHub repository](https://github.com/yandex-cloud-examples/yc-datasphere-jobs-examples) with job run examples