For reliable large-file upload tests in Playwright, handle three things separately: select the file, allow enough time for the application to process it, and assert that the application reports completion. Playwright can populate a file input; that alone does not prove the server accepted, stored, or processed the dataset.
Select the file through the right Playwright API
For a standard file input
Use a locator for the file input and call setInputFiles(). An accessible label is a good starting point:
As an Amazon Associate I earn from qualifying purchases.
const upload = page.getByLabel('Upload file');
await upload.setInputFiles('tests/fixtures/dataset.csv');Playwright’s file-upload documentation supports file paths, arrays of paths, and in-memory file objects containing a name, MIME type, and buffer. An array lets a test exercise multi-file selection:
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minuteawait page.getByLabel('Upload files').setInputFiles([
'tests/fixtures/part-1.csv',
'tests/fixtures/part-2.csv',
]);For a small generated fixture, an in-memory file object can avoid managing a fixture file on disk. For a genuinely large dataset, prefer a representative on-disk fixture: loading the entire dataset into a buffer in the test process can increase memory use. If the control supports directory selection, setInputFiles() also accepts a directory path. Pass an empty array to clear the selected files.
#1 Best Overall
For a file chooser opened by an action
When the application creates or opens the file input after a click, start waiting for the chooser before clicking. This prevents the test from missing the event:
const chooserPromise = page.waitForEvent('filechooser');
await page.getByRole('button', { name: 'Choose file' }).click();
const chooser = await chooserPromise;
await chooser.setFiles('tests/fixtures/dataset.csv');Use an accessible locator and the locator-based API where possible. The Locator API reference says setInputFiles() has no timeout by default unless a default or action timeout is configured. That does not remove the test runner’s separate overall timeout.
Give processing enough time without masking failures
File selection and application processing are different stages. A long-running upload test needs an overall test timeout that covers the expected end-to-end work, as well as an assertion timeout long enough for the completion signal to appear. Playwright Test’s documented defaults are 30,000 ms for a test and 5,000 ms for an assertion; these are defaults, not recommended budgets for a large upload. Changing one does not automatically change the other, as explained in Playwright’s timeout documentation.
For example, if the application normally needs up to two minutes to finish processing in the test environment, set a test budget that can accommodate that work and configure the relevant assertion’s timeout intentionally. The appropriate values depend on the application and environment; the documentation does not prescribe a large-upload budget.
Prefer a retrying assertion on the application’s real completion indicator over a fixed sleep. A larger timeout only gives the test more time; it does not make the check meaningful or establish that the upload succeeded.
Assert completion at the application boundary
After selecting the file, wait for a user-visible or server-confirmed result that represents the outcome you care about. For example, an application might display a completed status, show a success notification, or update a dataset record. Choose the signal that best confirms the workflow under test, and use a Playwright web assertion so the condition is retried until it passes or times out.
await expect(page.getByRole('status'))
.toHaveText('Upload complete', { timeout: 120_000 });This example assumes the application exposes a status element with that text; replace it with the actual completion signal and expected timing for your app. A populated file input confirms selection, not server acceptance or processing. Likewise, a click completing is not evidence that the dataset is ready.
Do not use networkidle as a generic upload-readiness test. Playwright’s API reference quotes its guidance as: “Don’t use this method for testing, rely on web assertions to assess readiness instead.” An application can have no active network requests before processing is complete, or keep requests open for unrelated reasons; assert the actual result instead.
Build a reliability matrix around the application’s policy
Playwright does not prescribe a benchmark suite for large uploads. A practical test matrix should reflect the application’s documented upload rules and exercise realistic workload variation:
- Vary dataset size, including representative boundary cases allowed by the application’s own upload policy.
- Vary the number of files and the formats the application supports.
- Record the runtime environment and observe whether the application reaches its completion state.
- Check successful uploads, validation rejections, and recoverable failure states.
- Prepare representative large fixtures on disk and plan their cleanup as part of the test design.
Compare like with like: note upload size, file count, runtime environment, and observed application behavior. These checks help distinguish selection problems from application validation, transfer, or processing failures; they are testing recommendations, not a Playwright-defined performance standard.
What Playwright does—and does not—guarantee
The cited Playwright documentation describes how to set files on an input and how to configure timeouts. It specifies neither a maximum upload size nor a throughput guarantee. Do not treat any particular dataset size as a Playwright limit, or promise a transfer speed based on the API alone. Upload capacity and runtime must be established by testing the target application and its environment.
The stable upload and timeout documentation supports the core workflow above, while some API links use the version-sensitive /docs/next/ path. Confirm API details against the Playwright version installed in your project.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
