TestMu AI, formerly LambdaTest, announced a Comparison view in App Profiling Insights that lets teams select multiple runs of the same mobile test and overlay them run by run. Announced on September 23, 2026 in San Francisco and Noida, the new view enables engineering teams to trace mobile performance regressions to device, OS version or build rather than just detecting that a regression happened.
Mobile performance regressions rarely announce where they came from. The new Comparison view removes guesswork by plotting individual runs against one another, filtered and grouped by device, OS version and app build, instead of averaging sessions into a single trend line.
The Comparison view removes guesswork from mobile performance debugging. Every selected run starts at 0:00 on a shared elapsed-time axis, so sessions of different lengths and start times line up and can be read against each other. One run is set as the baseline, and every other run's numbers show a colour-coded delta against it, green where the run is better, red where it is worse.
Because the same run keeps the same colour across every chart, a spike in memory can be followed straight down the page to see what frame rate and startup time were doing at the same moment.
Overlay up to five runs at once, across CPU, memory, frame rate, disk, network and startup time, with organisation-level SLA bands drawn on every chart. Filter and group by device, OS version, app build and pass/fail status, so a comparison can be narrowed to exactly the runs that isolate one variable.
Key capabilities include Baseline and Baseline Diff with per-metric deltas against any run chosen as reference, and Avg, Min, Max, and P90 side by side for every selected run, so a build that looks healthy on average but breaches on its worst runs is caught immediately.
Shared timeline zoom allows dragging a slice of one run to zoom every chart to the same window. Auto-saved comparisons restore selected runs, baseline and filters per test without a save button. The view is available now in App Profiling Insights for Appium tests on iOS and Android, alongside existing Trends view and Average/p90 toggle.
"Averages are how performance regressions hide. A build can look green on the dashboard and still be the one your users complain about, because the problem lives in one device or one OS version, and the mean smooths it away," said Mudit Singh, Co-Founder, TestMu AI. "Comparison puts the individual runs next to each other with the delta already computed. The question stops being 'did something regress?' and becomes 'which build, which device, which OS,' and the view answers that on the same screen."
The update enables teams to identify whether a CPU jump or slower cold start came from a new build, a newer OS version, or one specific device model behaving differently.
About TestMu AI
TestMu AI is the world's first Agentic AI-powered Quality Engineering platform, designed to enable organizations to automate and scale testing with intelligence at its core. By combining autonomous capabilities with seamless integration across modern development workflows, TestMu AI empowers teams to deliver faster, more reliable, and secure software in an AI-first world.
