ZeuZ
Performance testing

Performance testing tools that turn your API tests into load tests.

Performance testing tools in ZeuZ reuse the request steps from your API tests as load tests: set the spawn rate, the timeout and the sequence, run it from the pipeline and read throughput, latency percentiles, concurrency and error rate on the same report as everything else. Regressions in response time surface before users find them.

ThroughputP95 latencyConcurrencyError rateSoakFrom your CI
node-toronto-01atlas-crmrelease/2026.09
Load test · checkout API · 200 users · 10 min
Example run
Requests/s
17.2
Avg
1,047 ms
P95
2,105 ms
Success
100%
Peak conc.
64
Response time over runavgP95
0 min5 min10 min
Status codes
200 OK10,318
201 Created2,064
4xx0
5xx0
Built from the same request steps as the API testsrelease/2026.09
In one paragraph

Performance testing measures how an application behaves under load: throughput in requests per second, latency at the average and the 95th percentile, the concurrency it sustains and the error rate as demand rises, over a burst or a soak. ZeuZ runs load tests from the same request steps as its API tests, on your own nodes, and reports the numbers beside the functional results for the same build.

Part of ZeuZ Test Studio · APIs under load; desktop and web screen timings from functional runs
The problem

Where performance testing breaks today.

The load test is a separate tool with the same endpoints rewritten a second time, so it is out of date by the second sprint.

It runs before a launch and never again, so the regression that doubled P95 in March is found by customers in April.

The report is a chart in someone's folder; nobody can tell which build it measured.

The six steps

How ZeuZ automates performance testing.

  1. 01Test designProposes which endpoints and journeys need a load profile from the changed services and the traffic they carry.
  2. 02Test case creationDescribe the profile in plain English, "200 users for ten minutes against checkout"; ZeuZ builds the load test from the request steps you already have.
  3. 03Test automationLorust load actions compile the sequence with spawn rate, timeouts and the GET, POST, PUT, DELETE and GraphQL requests it should exercise.
  4. 04Test executionRuns from a CI webhook, nightly or manually, on your own nodes, so the load comes from inside your network when it has to.
  5. 05Reporting and analysisRequests per second, average and P95 latency, success and error rates, status codes, concurrency and bytes per second, pinned to the build.
  6. 06Test maintenanceWhen an endpoint changes, the API test and the load test change together because they are the same steps.

Your team reviews what ZeuZ produced. Nothing joins the suite or runs without a person accepting it.

See it in the demo lab or in your own environment

See it run on your application.

Watch an existing functional test turn into a load test, live.

What is covered

Load testing tools built from the tests you already have.

ZeuZ performance actions run over the Python requests driver and the Lorust load framework. A Lorust performance test sets the spawn rate for virtual users, the timeout and the action sequence, and a load test escalates request rates over a period to find the point where the service degrades. The request actions cover GET, POST with JSON, plain text, form-data, URL-encoded, empty and binary bodies, GraphQL queries, PUT, DELETE and HEAD, so the same steps used for functional API tests carry the load, which is what makes these performance testing tools part of the suite rather than a second tool.

Reports capture requests per second, average and 95th-percentile response time, success rate, error rate and status-code distribution, peak and average concurrency, and bytes transferred per second, with latency percentiles. For desktop and web applications, step-level durations from functional runs give a response-time trend per screen per build. Tests run on your own nodes, on-premises or air-gapped.

Engine
Lorust load actions over the Python requests driver
Profiles
Spawn rate, timeout, escalating rate, action sequence
Requests
GET, POST (JSON, text, form, URL-encoded, binary), GraphQL, PUT, DELETE, HEAD
Metrics
Requests/s, avg and P95 latency, success and error rate, status codes, concurrency, bytes/s
UI timing
Step durations per screen from functional runs
Deployment
Your nodes, on-premises, air-gapped
BurstSoakRampEndpoint regressionRelease-day capacity check

Runs where your code is allowed to be.

SaaS

Hosted by ZeuZ, execution on your own machines.

On-premises

Point the client at your own server.

Air-gapped

Disconnected environments, scoped per engagement.

Runs inside Jenkins, GitHub Actions, Azure DevOps, CircleCI and Bitbucket; results into Jira, Slack and Teams.

Related capabilities

Works in the same suite as.

Automated regression testing

The suite that catches the functional break while the load test catches the slow one.

Automated regression testing →
Desktop application testing

Screen-level step durations that trend response time per build.

Desktop application testing →
API test automation

The request steps that define the load profile, asserted functionally first.

API test automation →

Read next: Exploring performance and load testing with AI →Related: See pricing →

FAQ

Questions about performance testing.

Ready to automate QA testing today?

See ZeuZ Test Studio write, run and repair tests against your own application code.

or explore our services →
Book a demo