No drama. Just Lama.
Stop writing tests.
Start describing them.
Lama is an AI QA agent that navigates your app, learns how it works, and tests it - all from a simple conversation.
Works with your existing test suite
How it works
Tell Lama what to test. It figures out the rest.
01
Talk to Lama
Describe what you want to test in plain English. Lama navigates your app in a real browser, clicks through the flow, and verifies the outcome. If it's unsure about something, it asks you, like any good tester would.
02
Lama learns your app
The more you use Lama, the smarter it gets. It builds a knowledge base of your application so the next time you ask it to test something related, it already has context. No re-explaining. Your QA knowledge compounds over time.
03
Generate tests, run them, ship
Once Lama understands the flow, ask it to write a test case or generate a full automation script in Playwright, Cypress, or Selenium. Run your tests directly from Lama, group them into suites, and execute them on demand — all in plain English.
Everything your QA team needs. Nothing it doesn't.
01
Plain English, real results
Describe any test scenario the way you’d explain it to a colleague. Lama interprets your intent, navigates your app, and delivers working test output — no scripting, no selectors, no boilerplate. The barrier between thinking of a test and having a test disappears.
02
A knowledge system that remembers
Lama builds a living map of your application as you work. It remembers pages, user flows, form behaviors, and edge cases — so you don’t have to repeat yourself. When your team asks Lama to test a flow it’s seen before, it starts with context, not from scratch. QA knowledge finally lives somewhere useful.
03
Real code, your framework
When you need automation scripts, Lama generates native code in Playwright, Cypress, or Selenium. No proprietary format. No lock-in. The tests run in your CI pipeline, live in your repo, and work whether or not you keep using Lama.
04
Connects to the tools you already use
Sync test cases to TestRail or Xray. Create and link tickets in Jira, Azure DevOps, or Asana. Lama integrates into your existing QA and project management workflow — it doesn’t replace it. Your team stays in their tools. Lama just makes them more productive.
05
Run and manage tests from one place
Execute individual tests or entire suites directly from Lama. Group tests by feature, sprint, or priority. Review results, track what’s passing, and see what needs attention — all without leaving the app. Think of it as the command center your QA workflow has been missing.
06
Desktop app + Chrome extension
Use the full desktop app on Mac or Windows for a dedicated testing workspace. Or install the Chrome extension to test on the fly while browsing your application. Both share the same AI agent and the same knowledge base — start in one, continue in the other.
Pricing
Plans for every team size.
No subscription required. Buy AI credits as you need them — add seats and integrations as your team grows.
Pay as you go
Pay per use
no subscriptionFor individual QA engineers exploring AI-powered testing.
Team
$79
seat / monthFor QA teams that need collaboration, integrations, and shared knowledge.
Enterprise
Custom
contact usFor organizations that need security, control, and dedicated onboarding.
All plans include pay-as-you-go AI credits. Credits are deducted per agent action and never expire.
What developers say
Built for teams that ship fast.
"
I described a five-step checkout flow in two sentences. Lama had a working Playwright test in under a minute.
Lamija D.
QA Engineer, early access
"
The knowledge system is the real unlock. After a week, Lama knew our app better than our documentation.
Adnan P.
QA Lead, early access
"
We evaluated three AI testing tools. Lama was the only one that didn’t try to lock us into a proprietary platform.
Harvey E.
Engineering Manager
FAQ
Common questions.
How does Lama actually test my app?
Lama launches a real browser, navigates to your application, and interacts with it — clicking, typing, scrolling — based on your plain English instructions. It reads the DOM to understand your UI, verifies outcomes, and builds a knowledge base it can reference in future sessions.
What can I do with the test output?
You can ask Lama to generate test cases (for manual QA or syncing to test management tools), or generate full automation scripts in Playwright, Cypress, or Selenium. The code is native to the framework — no proprietary dependencies. Commit it, run it in CI, treat it like any code your team writes.
Does Lama work with my existing test setup?
Yes. Lama generates standard framework code and integrates with tools like Jira, Azure DevOps, Asana, TestRail, and Xray. There’s nothing proprietary to install in your test infrastructure.
What does "knowledge system" mean?
As you use Lama, it builds an understanding of your application — the pages, flows, components, and behaviors it has observed. This knowledge persists across sessions and, on the Team plan, is shared across your team. It means Lama gets faster and more accurate the more you use it.
What platforms are supported?
The desktop app is available for Mac, Windows, and Linux. The Chrome extension works on any platform that supports Chrome. Both connect to the same AI agent and knowledge base.
How does pay-as-you-go pricing work?
You prepay for AI credits and use them as you go. Credits are consumed when Lama generates tests, executes them, or builds knowledge. This way, you only pay for what you actually use. No surprise invoices — you control your spending by topping up when you need more.
Is my app data private?
Yes. Your code and project files never leave your machine. AI conversations are encrypted locally. When you use the AI agent, prompts are sent through our backend to the AI provider (Anthropic) — they don’t train on your data, and we don’t store your prompts or responses. See our Data Use page for the full breakdown.
Can my whole team use Lama?
Yes. The Team plan includes shared workspaces and a shared knowledge base — so when one person teaches Lama about a flow, the whole team benefits. Credits are prepaid per team and deducted from a shared balance, but you can set up a monthly seat limit for each user.
Get started
Stop writing tests.
Start describing them.
Create an account, download the app, and let the agent handle the rest.