Complex testing. Simple pricing.
Pay for runs. Unlimited AI usage, 100% parallelization, and no user limits.
Invite your agent to the packPrice per run
A run is a complete QA cycle: selecting tests to run, parallel execution, clearing flakes, AI maintenance, and AI bug reproduction and reporting.
| Platform | Per run |
|---|---|
| WebChrome, Firefox, and WebKitElectron apps on Linux | $1 |
| MobileiOS on real iPhones and iPadsAndroid on emulators or real devices | $2 |
| DesktopWindows and macOS (native or Electron) | $2 |
| Session timeBrowser or device time outside a run (e.g. creating tests) | Every 5 minutes billed same as above |
Up to 5 minutes of run time per test. Longer tests are billed in 5-minute blocks. For example, a 7-minute web test is $2 per run. Each run is attempted up to 3 times for free to catch flakes and false positives. Volume pricing is available: talk to sales.
Unlimited AI included
No AI credits or token bills.
Coverage Planning
Agents explore your app and map the test cases it needs, so you know what to cover before a test is written.
Test Creation
Agents write the tests in open-source Playwright and Appium.
Smart Smoke Suites
Agents pick the tests each pull request needs and run only those.
Failure Investigation
Every failure is diagnosed by an agent before anyone on your team looks at it.
Test Maintenance
Broken tests are fixed automatically as your app changes.
Bug Reporting
When a test finds a bug, the agent reproduces it, then files it in Linear or Jira.
Pay for runs, not flakes
When QA Wolf runs your tests, you pay for verified pass/fail results that tell you it's safe to ship, not compute time and tokens.
Any of these count as one run:
Want us to do the work?
Jump Start
We write your tests.
QA Wolf's team builds your suite, then hands it to your team to run on the Platform.
Contact salesCoverage-as-a-Service
We handle it all.
QA Wolf's team writes, runs, and maintains your tests and investigates every failure. We guarantee 80%+ automated coverage, and you're never alerted to a flake.
Contact salesRuns
What counts as a test run?
One test on one platform, from start to a verified result. Everything it takes to get there is included.
How are runs billed?
Each 5 minutes of a run is billed at its platform's price, so a 7-minute web run is $2.
Do retries cost extra?
No. A failing test runs up to 3 times to clear flakes, and the run is billed once, on its longest attempt.
Does investigating a failure cost extra?
No. QA Wolf's agent investigates failures as part of the run.
Does reproducing a bug cost extra?
No. The agent re-runs the test to confirm the bug and files a report, as part of the run.
Does fixing a broken test cost extra?
No. The agent repairs tests that break in a run, as part of the run. Edits you ask for in the editor are session time.
Do parallel runs cost extra?
No. Tests run in parallel by default, or in the order you set with Run Rules.
Session time
What is session time?
Any time a browser or device is open outside a run, such as when you or the agent creates tests.
How is session time billed?
In 5-minute blocks at a run's price, added up for each platform over the month. Two 2-minute web sessions are one $1 block.
Why is session time billed separately from runs?
Runs are priced by result, so your bill is predictable and making the AI efficient is our job. Creating tests varies by task, so it's billed by time.
Do I have to pay for test creation?
Only the session time it takes. After that, you pay per run.
Am I billed while a browser sits idle?
For up to 5 minutes. After 5 minutes without activity, the browser closes.
Does reopening a browser start a new 5-minute block?
No. The time is added to the month's total for that platform.
Does messaging the agent open a browser?
No. The agent opens one only when it needs it, or when you run code.
Billing
Does QA Wolf charge per seat?
No. Add as many people as you like.
Do I pay for AI usage?
No. AI is unlimited. AI and session time each have a monthly fair-use limit.
Do I pay in advance?
Yes. You buy credit up front, and it tops up automatically when it runs low.
Are there volume discounts?
Yes. Let's talk.
Can I set a spending limit?
Yes. A billing manager can cap monthly top-ups. At the cap, new work pauses until it's raised or the month resets.
Platform
What platforms does QA Wolf support?
Web: Chrome, Firefox, and WebKit (Safari)
iOS: Real iPhones and iPads
Android: Real devices or emulators
Desktop: Windows and macOS (native or Electron), plus Electron on LinuxCan I bring my own Playwright tests?
Yes. QA Wolf runs vanilla Playwright.
Am I locked into QA Wolf's platform?
No. Your tests are Playwright and Appium code you can export anytime.
Services
What is Coverage-as-a-Service?
QA engineers on top of the platform who plan and create your tests and verify results. With Jump Start, they review and approve your suite, then hand it to your team.
How much do Jump Start and Coverage-as-a-Service cost?
Both are quoted based on the tests you need. Contact sales.
For developers
Introduce your agent to QA Wolf.
Add this server to Claude Code, Codex, Cursor or any MCP client, then ask for your first test.
Setup for each agentFor engineering leaders
See QA Wolf on your app.
Get a walkthrough on your own product and a plan for the coverage you need.
Contact sales- Send bugs to
- Your data
QA Wolf doesn’t train or fine-tune models on customer data. Privacy Policy