Do my existing Gatling simulations run without changes?

Yes. Gatling Java DSL simulations run on HyperExecute as they are, with no rewrite and no proprietary format to convert to. You declare the Java runtime version your simulation compiles against in the configuration, and the platform resolves dependencies before the run starts.

Is Gatling on TestMu AI Scala or Java?

Java. The supported path is the Gatling Java DSL, and the upload step asks for .java simulation files. Some documentation still describes Gatling as Scala based because of its origins, but the official sample used here contains no Scala. Check your simulation is on the Java DSL before you plan a migration.

Can I run Gatling from my CI pipeline?

Yes, and this is the reason to pick Gatling over JMeter on this platform. Gatling runs from a hyperexecute.yaml committed in your own repository, driven through the CLI, so the run configuration is versioned next to the simulation it belongs to. JMeter is portal only and has no CLI or YAML path.

How do I shape the load profile?

Through the simulation itself or through variables in the configuration. The sample configuration exposes an injection type, a workload model, a user count, a duration, a ramp duration, and start and end user counts as variables that the simulation reads as system properties. Load can also be shaped from the portal form.

What results do I get from a Gatling run?

The Gatling HTML report, collected as a downloadable job artifact once the run completes, alongside the run logs. This is Gatling's own report rather than a separate reporting layer, so the output looks the way your team already reads it. The JMeter path on this platform produces a different set of five dashboard reports instead.

Where is load generated from?

Six named regions are documented: East US in Richmond, Virginia, which is the default, West US 2 in Moses Lake, Washington, Central India in Pune, Southeast Asia in Singapore, Brazil South in Sao Paulo State, and Mexico Central in Queretaro State. Each is a named physical location rather than an unspecified count of zones. Multi region load generation is an Enterprise plan entitlement; Free, Basic and Pro run single region.

How many virtual users can a Gatling run reach?

Concurrency is a plan entitlement: 100 concurrent virtual users on Free, up to 2,000 on Basic, up to 8,000 on Pro, and a custom figure on Enterprise. The real figure inside that ceiling depends on the work each virtual user performs. Browser-driving samplers reduce it, as do missing HTTP timeouts, large payloads, and long chained workflows. Size a run against that behaviour rather than against a headline number.

Should my team standardise on Gatling or JMeter?

Pick Gatling if you want load as code, run configuration versioned in git, and a pipeline-triggered load stage. Pick JMeter if your team already maintains large .jmx plans, prefers GUI authoring, and is happy configuring runs through a portal form. Both formats are supported natively here, so the decision is about how your team works rather than what the platform allows.

Does this cover browser performance?

No. Gatling here is protocol level load testing, which measures what your services return under concurrent request volume. For rendering behaviour and front-end timing, run Playwright or Selenium on the same platform instead.

Do I still need my own load generator infrastructure?

No. HyperExecute provisions the generators when a run begins and destroys them when it ends, so nothing sits idle between tests and there is no fleet to patch. Concurrency comes from your plan pool.

Gatling Cloud Load Testing

Run the Gatling Java DSL simulations you already have. Run configuration lives in your repository, load generators are provisioned per run and destroyed when it ends.

Start free with Google

Start free with Email

Automate Browser Flows from your
Terminal with Kane CLI

Explore Kane CLI

Trusted by 3M+ users globally at

+Read case study

Load as Code, Without the Load Generators

Gatling simulations run unchanged on HyperExecute, driven from the portal or from a config file in your repo.

Start free with Google

NO REWRITE

Run Simulations Unchanged

Existing Gatling Java DSL simulations run as they are. The platform resolves dependencies, drives the run, and reports.

Start free

Java DSL simulations run with no rewrite required.
Gatling HTML report collected as a job artifact.
Generators destroyed automatically when the run ends.

CONFIG IN GIT

Keep Run Config in Your Repo

Gatling is the format that gets both the portal and the hyperexecute.yaml path, so config is versioned with the simulation.

Read the setup docs

Run configuration reviewed in the same pull request.
Injection type, users, duration and ramp set as variables.
Declare the Java runtime the simulation compiles against.

SIX REGIONS

Choose Where Load Comes From

Six named regions, each a named physical site, weighted by percentage so a run matches where your users are.

Start free

East US, West US 2, Central India and Southeast Asia.
Brazil South and Mexico Central for Latin America.
Traffic split by percentage rather than fixed per region.

Where our own coverage is uneven

Gatling Gets Both Paths. JMeter Gets One.

We support both plan formats natively, but not identically, and the difference decides which one a pipeline-driven team should choose. Gatling runs from the portal and from a hyperexecute.yaml in your repository. JMeter runs from the portal only, with no CLI or YAML path. We would rather say that here than let you discover it in week one.

Versioned Together

The simulation and the configuration that runs it live in the same repository and move through the same review.

Triggered From CI

A load stage runs from the CLI like any other pipeline step, so a load run belongs to a commit rather than to a person.

Shaped by Variables

Injection type, workload model, user count, duration and ramp are variables the simulation reads, not fields someone retypes.

On capacity, your ceiling is the one your plan entitles: 100 concurrent virtual users on Free, up to 2,000 on Basic, up to 8,000 on Pro. Inside that, browser-driving samplers reduce what you achieve, as do missing HTTP timeouts, large payloads, and long chained workflows. Size the run against what each virtual user actually does. If you maintain .jmx plans instead, the JMeter cloud load testing path covers that format.

Open Source Gatling and Gatling Here

Start free with Google

The simulation is identical. What changes is who provisions, drives and destroys the machines that generate the load.

Gatling on your own infrastructure

Gatling on TestMu AI

Simulation format

Java DSL

Java DSL, unchanged

Load generators

You provision and patch them

Provisioned per run, destroyed after

Between runs

Fleet idles or is rebuilt

Nothing is running

Multi-region load

You build and coordinate it

Six named regions, split by percentage

Run configuration

Scripts you maintain

hyperexecute.yaml in your repo

Reporting

Gatling HTML you collect

Gatling HTML collected as a job artifact

Functional suites

A separate system

Same grid as Selenium and Playwright

Load Runs Beside Everything Else You Test

One grid, one account and one bill across load plans, functional suites and real device runs.

HyperExecute

Automation Cloud

KaneAI

Test Intelligence

HyperExecute

The orchestration cloud that runs Gatling simulations, provisions the generators, and destroys them afterwards.

More about HyperExecute

Config in Git

A hyperexecute.yaml sits beside the simulation in your own repository.

Java Runtime

Declare the language and version your simulation compiles against.

Job Artifacts

The Gatling HTML report is collected against the job on completion.

Performance Testing Plans and Limits

Start free with Google

Concurrency, virtual user hours, load generators and test duration are all plan entitlements. Verified on the pricing page on 8 September 2026.

Free

Basic

Pro

Enterprise

Price, annual

$79 / month

$349 / month

Custom

Price, monthly

$99 / month

$449 / month

Custom

Concurrent virtual users

100

Up to 2,000

Up to 8,000

Customizable

Virtual user hours / month

100, introductory

2,000, introductory

8,000

Customizable

Load generators

Up to 2

Up to 20

Customizable

Max test duration

40 minutes

2 hours

6 hours

Customizable

Load regions

Single region

Multi region

Some Love from our Customers

As Best Egg expanded its product offerings and entered new markets, we knew our old testing infrastructure couldn’t keep up.
With support from Tenny Agustin, our Engineering Operations Lead, we modernized our approach with

TestMu AI

Best Egg

best-egg

Excited to Share My Learning Journey with Kane AI & Lambda Tool!
I'm pleased to announce that I've recently gained hands-on experience exploring Kane AI through the Lambda Tool and it’s been a fantastic journey of upskilling!

KaneAI

Suryateja Goud

suryateja-goud

See how is #Futureready to enable blazing-fast test orchestration seamlessly integrated with organizations' existing CI/CD platforms, using #Microsoft Azure.

TestMu AI