Holmes vs QA Wolf
Both use AI agents to plan, run and repair end-to-end tests, and both price on usage with no seat limits. QA Wolf is strong on breadth: real devices, desktop apps and a managed service. Holmes adds agents that explore paths nobody wrote down, context from specs, designs and analytics, and a summary of every release.
QA Wolf runs fixed Playwright and Appium code by design, and its agents explore your app to plan coverage. Holmes agents explore your product while they test and check paths nobody wrote down.
QA Wolf's agents take context your coding agent passes over MCP, plus docs and test plans you give them. Holmes reads the code change, the ticket, the spec in Notion or Confluence, the design in Figma and your analytics in PostHog before it decides what to test.
When something ships, Holmes writes what changed in plain language, then tests the journeys the change touched, so the whole team knows what went out.
Holmes is a Belgian company and GDPR compliant. If where your test data is processed matters to you, compare Holmes' Trust Center with QA Wolf's privacy policy, which covers processing in the United States.
Based on each product's public pages on 17 November 2026. A red cross means QA Wolf's pages don't describe it, not that it can't be done.
| Capability | Holmes | QA Wolf |
|---|---|---|
| Testing | ||
| How tests are written | Yes: Plain-language goals. No test code to write | Yes: Agents write Playwright or Appium code from a prompt or a recording; you can export it[1][2] |
| Finds bugs nobody scripted | Yes: Agents explore and test paths nobody wrote down | Partly: Agents explore to plan coverage; runs replay fixed Playwright code by design[2][3] |
| Keeps tests working when the UI changes | Yes: Flows repair themselves when a label or layout changes | Yes: An agent repairs broken tests and keeps the same assertions[1] |
| Other browsers | Yes: Chromium, Firefox and WebKit, the engine behind Safari | Yes: Chrome, Firefox and WebKit (Safari), plus Electron[2] |
| Native Android apps | Yes: Native Android apps on an Android emulator | Yes: On real devices or emulators[2] |
| Native iOS apps | Yes: Native iOS apps on iPhone and iPad simulators | Yes: On real iPhones and iPads[2] |
| Real devices | Not covered: Not covered. Android emulators and iOS simulators | Yes: 500+ device and OS combinations on demand[4] |
| APIs | Yes: Every request behind a journey, checked in context | Partly: API calls inside end-to-end tests; a standalone API suite isn't described[5] |
| Accessibility | Yes: WCAG audits in a real browser, with keyboard navigation | Yes: WCAG and Section 508 scans, with audit logs[6] |
| Visual testing | Partly: Compares the Figma design with what shipped. No pixel-diff baselines | Yes: Key screens compared with an approved baseline[6] |
| Workflow | ||
| Pull requests and previews | Yes: Check and comment on GitHub and Bitbucket pull requests. GitLab supported | Yes: Agents pick the tests each pull request needs; runs on preview environments[2][1] |
| Context it reads before testing | Yes: Reads the code change, ticket, spec (Notion, Confluence), design (Figma) and analytics (PostHog) | Partly: Context your coding agent passes over MCP, plus docs or test plans you give it. Figma, Notion or analytics aren't on its pages[3] |
| Coding agents (MCP) | Yes: Start runs and read findings from Claude and Cursor via MCP | Yes: MCP server for Claude Code, Codex and Cursor[6] |
| Evidence with each bug | Yes: Recording, screenshots, console and network logs, steps to reproduce | Yes: Video, repro steps, traces and logs[3] |
| Where findings go | Yes: Findings on the pull request, the ticket, Slack, Microsoft Teams and email | Yes: Bugs filed in Linear or Jira; ask from Slack, Teams, GitHub or GitLab[2][6] |
| Pricing and security | ||
| What you pay for | Yes: Priced on how much Holmes tests, never on seats | Note: Per test run of up to 5 minutes: $1 web, $2 mobile or desktop. No user limits[2] |
| Managed service | Not covered: Not covered. Self-serve, with guided onboarding on Team and up | Yes: Coverage-as-a-Service: QA Wolf's engineers write, run and maintain tests; quoted by sales[6][2] |
| Getting started | Yes: €250 in free credits, no credit card | Yes: Free runs for every new team[2] |
| Where data is processed | Yes: EU company (Ghent, Belgium), GDPR compliant | Note: Anywhere in the world, including the United States, per its privacy policy[7] |
Holmes sells QA software, so we compete with QA Wolf. We've tried to be fair: every statement about QA Wolf comes from its own public pages, linked below with the date we checked them. Products change. If something here is out of date, email support@useholmes.com and we'll fix it.
From a prompt: Simply describe the flow you want tested and the agent will get to work. After walking through the flow manually, it’ll write deterministic Playwright code.Checked 17 November 2026.
Pay per test run. Unlimited AI usage, 100% parallelization, and no user limits.Checked 17 November 2026.
QA Wolf runs Playwright and Appium tests instead of relying on computer-use agents that never run the same flow twice.Checked 17 November 2026.
device and OS combinations on demandChecked 17 November 2026.
Call your API after a step to check the change saved, not just showed.Checked 17 November 2026.
The most advanced agents, infrastructure, and experts for creating, running, and maintaining E2E tests.Checked 17 November 2026.
may be transferred, processed, and stored anywhere in the world, including, but not limited to, the United StatesChecked 17 November 2026.
QA Wolf builds deterministic Playwright and Appium suites with agents, and offers a managed service where its own engineers write and maintain them. It covers real devices and desktop apps. Holmes tests in plain language with agents that explore, reads specs, designs and analytics, and writes a summary of every release.[6][2][1]
With QA Wolf, yes: its tests are open-source Playwright and Appium code you can export at any time. Holmes works from plain-language goals, so there is no test code to write or maintain. You can import existing test cases from a file or your tracker to get started.[2]
QA Wolf tests iOS on real iPhones and iPads and Android on real devices or emulators, at $2 a run. Holmes tests native Android and iOS apps on emulators and simulators; real devices are not covered.[2]
Both price on usage and neither charges per seat. QA Wolf charges $1 per web test run and $2 per mobile or desktop run, for up to 5 minutes each, with free runs for new teams; its managed plans are quoted by sales. Holmes has a Free plan, Starter at €199 a month for up to 3 products, and Team from €499 a month, all with unlimited users.[2]
QA Wolf offers Coverage-as-a-Service: its team writes, runs and maintains your tests, investigates every failure within 24 hours and guarantees 80%+ automated coverage. Holmes is self-serve, with guided onboarding on Team and up.[6][2]
If you need testing on real iPhones and Android phones, Windows or macOS desktop apps, a team that writes and maintains your suite for you, or exportable Playwright code, QA Wolf covers those on its own pages. If you want agents that explore your product with context from your specs, designs and analytics, from an EU company, try Holmes.[2][6][1]
Evaluating QA Wolf? See what Holmes finds first.
Free, no scripts, no SDK, and no test account needed to start.