Question 1

Is TestRelic an AWS Device Farm alternative?

Accepted Answer

It can be, but it isn't only that. Studio's Infrastructure pane provisions real machines directly in your own AWS account the same way AWS Device Farm does — Studio creates the network and the machine, installs the browsers, and connects it, so you only pick a size. The difference is you aren't limited to AWS: the same pane also connects a hosted device grid (BrowserStack or TestMu AI) or just uses your own computer, and every session — local or cloud — runs against whichever target you picked, from one place instead of three separate accounts.

Question 2

Can I run Playwright tests on my own AWS, Azure, or GCP account?

Accepted Answer

Yes. Studio creates the network, machine, and browsers in your own account and connects it — you only pick a size. On AWS that ranges from a t3.medium (2 vCPU, 4 GB, $0.0416/hour) for a small suite up to a c6i.4xlarge (16 vCPU, 32 GB, $0.68/hour) for nightly regression, with Azure and GCP offering equivalent tiers. Studio only ever creates a private network and machines tagged for it — no IAM users, no changes to anything else already in your account — and the machine accepts no inbound traffic until a run actually starts.

Question 3

Does TestRelic integrate with BrowserStack and TestMu AI (formerly LambdaTest)?

Accepted Answer

Yes, both as connectable targets in Studio's Infrastructure pane and, separately, as Cloud Farms integrations on the platform for pulling in build data. LambdaTest rebranded to TestMu AI in January 2026 — same product, same API, same account, new name. Once connected, a target from either shows real browsers and real devices as a hosted grid option; TestMu AI's grid additionally heals flaky selectors while a run is in progress.

Question 4

What's the difference between Local mode and Cloud mode infrastructure?

Accepted Answer

Each mode has its own default target, set independently. Local mode defaults to "This computer" — nothing provisioned, results never leave your machine. Cloud mode defaults to "TestRelic Cloud" — results upload and sync to your team. Either default can be swapped for a provisioned AWS/Azure/GCP machine or a hosted device grid; the choice is made once per mode in Studio's Infrastructure pane and every session in that mode runs against it, though a single run can still target something else from the chat box without changing the default.

Question 5

How much does it cost to run tests on my own cloud account through TestRelic?

Accepted Answer

Two separate costs. TestRelic itself charges nothing extra for Infrastructure beyond the Studio-inclusive Pro plan ($99/month, $79/month billed annually) — provisioning, connecting, and switching targets is part of that plan. Your cloud provider bills you directly and only for what actually runs: an AWS t3.large used for testing runs about $0.0832/hour, so a team running 4 test-hours a day lands around $7-8/month in AWS spend. There is no markup and no TestRelic-managed compute in between.

Question 6

Do I need to provision infrastructure before I can run tests?

Accepted Answer

No. "This computer" and "TestRelic Cloud" are built-in defaults, already selected for Local and Cloud mode respectively — nothing to provision, nothing to connect. You only open the Infrastructure pane and add a target when a suite outgrows a laptop: more parallel runners than one machine can give you, or real devices you don't personally own. Most teams run entirely on the two defaults for weeks before their first AWS machine or BrowserStack connection.

Where your tests actually run.

Infrastructure

One infrastructure target, picked once

Two defaults, zero setup

Your own AWS, Azure, or GCP

A hosted device grid

Chosen once, from Studio

Every target, at a glance

Built in

Your own cloud

Hosted device grid

The same catalog, in the app

Built in

Your own cloud

A hosted device grid

Common questions

Stop hard-coding where your tests run.