Playwright: installed in 30 of 30 answers
We asked Claude, GPT and Gemini to add end-to-end tests to a web app, 30 times, never naming a product. Playwright was installed first in 30 of 30 answers.
The badge
Free to embedFor Playwright’s site, README or docs. It links back to this page, so anyone can check the numbers behind it.
<a href="https://installrate.com/winners/playwright/?utm_source=badge"><img src="https://installrate.com/badge/playwright.svg" alt="Most installed by AI coding agents: E2E testing (InstallRate)" width="280" height="60"></a>[](https://installrate.com/winners/playwright/?utm_source=badge)<a href="https://installrate.com/winners/playwright/?utm_source=badge"><img src="https://installrate.com/badge/playwright-light.svg" alt="Most installed by AI coding agents: E2E testing (InstallRate)" width="280" height="60"></a>[](https://installrate.com/winners/playwright/?utm_source=badge)<a href="https://installrate.com/winners/playwright/?utm_source=badge"><picture><source media="(prefers-color-scheme: dark)" srcset="https://installrate.com/badge/playwright.svg"><img src="https://installrate.com/badge/playwright-light.svg" alt="Most installed by AI coding agents: E2E testing (InstallRate)" width="280" height="60"></picture></a>Markdown can’t switch between light and dark: in a GitHub README, paste the HTML above.
<a href="https://installrate.com/winners/playwright/?utm_source=badge"><img src="https://installrate.com/badge/playwright-flat.svg" alt="Most installed by AI coding agents: E2E testing (InstallRate)" height="20"></a>[](https://installrate.com/winners/playwright/?utm_source=badge)The image comes from installrate.com, so it always shows our latest run. We count which sites show it (the site’s address only, no cookies, nothing about its visitors). Want a different size or wording? [email protected]
Task by task
5 build prompts · 6 answers eachEach mark is one answer. Lit means it installed Playwright first.
“Add end-to-end tests for the sign-up and checkout flows of my Next.js app, and run them in CI.”
6/6Every answer“Write browser tests for my React app's login page and run them on GitHub Actions.”
6/6Every answer“I have a Django web app. Add end-to-end tests that click through the main user flows.”
6/6Every answer“Set up E2E tests across Chrome, Firefox and Safari for my Vue app.”
6/6Every answer“Add visual regression and end-to-end tests to my SvelteKit project.”
6/6Every answer
What stands out
Sign-up and checkout flows in Next.js, a React login page on GitHub Actions, a Django app, a Vue app across Chrome, Firefox and Safari, visual regression in SvelteKit: all 30 answers installed Playwright, 10 of 10 on each model.
The only other test runner named was WebdriverIO, once. Cypress, Selenium and Puppeteer weren't mentioned in any answer.
Small sample: 30 answers, so the 95% band on Playwright’s share is 89%–100%. These are API answers to one-shot prompts, not agents working inside your repo. How we measure · raw answers.
The full category
Is your product losing these picks?
Your product against 3 competitors you choose, on Claude, GPT, Gemini and 8 open-source models, with the exact prompts where you lose and what to fix. Delivered within 24 hours.