> actions/upload-artifact@v3 is scheduled for deprecation on **November 30, 2024**. [Learn more.](https://github.blog/changelog/2024-04-16-deprecation-notice-v3-of-the-artifact-actions/)
> Similarly, v1/v2 are scheduled for deprecation on **June 30, 2024**.
> Please update your workflow to use v4 of the artifact actions.
> This deprecation will not impact any existing versions of GitHub Enterprise Server being used by customers.
Upload [Actions Artifacts](https://docs.github.com/en/actions/using-workflows/storing-workflow-data-as-artifacts) from your Workflow Runs. Internally powered by [@actions/artifact](https://github.com/actions/toolkit/tree/main/packages/artifact) package.
- [Where does the upload go?](#where-does-the-upload-go)
## v4 - What's new
> [!IMPORTANT]
> upload-artifact@v4+ is not currently supported on GHES yet. If you are on GHES, you must use [v3](https://github.com/actions/upload-artifact/releases/tag/v3).
The release of upload-artifact@v4 and download-artifact@v4 are major changes to the backend architecture of Artifacts. They have numerous performance and behavioral improvements.
For more information, see the [`@actions/artifact`](https://github.com/actions/toolkit/tree/main/packages/artifact) documentation.
1. Uploads are significantly faster, upwards of 90% improvement in worst case scenarios.
2. Once uploaded, an Artifact ID is returned and Artifacts are immediately available in the UI and [REST API](https://docs.github.com/en/rest/actions/artifacts). Previously, you would have to wait for the run to be completed before an ID was available or any APIs could be utilized.
3. The contents of an Artifact are uploaded together into an _immutable_ archive. They cannot be altered by subsequent jobs unless the Artifacts are deleted and recreated (where they will have a new ID). Both of these factors help reduce the possibility of accidentally corrupting Artifact files.
1. On self hosted runners, additional [firewall rules](https://github.com/actions/toolkit/tree/main/packages/artifact#breaking-changes) may be required.
2. Uploading to the same named Artifact multiple times.
Due to how Artifacts are created in this new version, it is no longer possible to upload to the same named Artifact multiple times. You must either split the uploads into multiple Artifacts with different names, or only upload once. Otherwise you _will_ encounter an error.
| `artifact-id` | GitHub ID of the Artifact, can be used by the REST API | `1234` |
| `artifact-url` | URL to download the Artifact. Can be used in many scenarios such as linking to artifacts in issues or pull requests. Users must be logged-in in order for this URL to work. This URL is valid as long as the artifact has not expired and the artifact, run, and repository have not been deleted. | `https://github.com/example-org/example-repo/actions/runs/1/artifacts/1234` |
For supported wildcards along with behavior and documentation, see [@actions/glob](https://github.com/actions/toolkit/tree/main/packages/glob) which is used internally to search for files.
If multiple paths are provided as input, the least common ancestor of all the search paths will be used as the root directory of the artifact. Exclude paths do not affect the directory structure.
Relative and absolute file paths are both allowed. Relative paths are rooted against the current working directory. Paths that begin with a wildcard character should be quoted to avoid being interpreted as YAML aliases.
If you are uploading large or easily compressable data to your artifact, you may benefit from tweaking the compression level. By default, the compression level is `6`, the same as GNU Gzip.
The value can range from 0 to 9:
- 0: No compression
- 1: Best speed
- 6: Default compression (same as GNU Gzip)
- 9: Best compression
Higher levels will result in better compression, but will take longer to complete.
For large files that are not easily compressed, a value of `0` is recommended for significantly faster uploads.
For instance, if you are uploading random binary data, you can save a lot of time by opting out of compression completely, since it won't benefit:
But, if you are uploading data that is easily compressed (like plaintext, code, etc) you can save space and cost by having a higher compression level. But this will be heavier on the CPU therefore slower to upload:
```yaml
- name: Make a file with a lot of repeated text
run: |
for i in {1..100000}; do echo -n 'foobar' >> foobar.txt; done
If a path (or paths), result in no files being found for the artifact, the action will succeed but print out a warning. In certain scenarios it may be desirable to fail the action or suppress the warning. The `if-no-files-found` option allows you to customize the behavior of the action if no files are found:
In matrix scenarios, be careful to not accidentally upload to the same artifact, or else you will encounter conflict errors. It would be best to name the artifact _with_ a prefix or suffix from the matrix:
Previously the behavior _allowed_ for the artifact names to be the same which resulted in unexpected mutations and accidental corruption. Artifacts created by upload-artifact@v4 are immutable.
Environment variables along with context expressions can also be used for input. For documentation see [context and expression syntax](https://help.github.com/en/actions/reference/context-and-expression-syntax-for-github-actions):
The retention period must be between 1 and 90 inclusive. For more information see [artifact and log retention policies](https://docs.github.com/en/free-pro-team@latest/actions/reference/usage-limits-billing-and-administration#artifact-and-log-retention-policy).
If an artifact upload is successful then an `artifact-id` output is available. This ID is a unique identifier that can be used with [Artifact REST APIs](https://docs.github.com/en/rest/actions/artifacts).
Although it's not possible to mutate an Artifact, can completely overwrite one. But do note that this will give the Artifact a new ID, the previous one will no longer exist:
You may also be limited by Artifacts if you have exceeded your shared storage quota. Storage is calculated every 6-12 hours. See [the documentation](https://docs.github.com/en/billing/managing-billing-for-github-actions/about-billing-for-github-actions#calculating-minute-and-storage-spending) for more info.
When an Artifact is uploaded, all the files are assembled into an immutable Zip archive. There is currently no way to download artifacts in a format other than Zip or to download individual artifact contents.
File permissions are not maintained during artifact upload. All directories will have `755` and all files will have `644`. For example, if you make a file executable using `chmod` and then upload that file, post-download the file is no longer guaranteed to be set as an executable.
If you must preserve permissions, you can `tar` all of your files together before artifact upload. Post download, the `tar` file will maintain file permissions and case sensitivity.