Momentic review

Momentic is a hosted software testing platform with a CLI and test execution infrastructure, not a standalone testing library. Its Mo agent explores web, iOS, and Android apps, files bugs with reproduction steps and recordings, and can turn an approved flow into a YAML regression test. That puts it alongside a team’s existing application code and CI pipeline: tests can live in a repo and run on pull requests, schedules, or through CI.

For web testing, Momentic lists Chromium and Chrome on local or hosted browsers. For mobile, it supports iOS simulators and Android emulators, either local or Momentic-hosted. The homepage also says tests can run on your own machines, in CI, or on Momentic-hosted infrastructure. This is useful if you want exploratory testing and repeatable checks in one workflow. It does not replace the application’s build system or CI provider; it supplies the test authoring and execution layer.

The homepage names a CLI for authoring and running tests from a terminal, plus an MCP server that lets a coding agent write, run, and debug tests. It also lists API keys, a test format, CLI setup, MCP server setup, and CI guides in its developer resources. Tests use YAML files in the repo, and the platform says approved flows can be submitted as pull requests.

CI integration is described for GitHub Actions, GitLab, or any CI provider, with runs on every push, pull request, or schedule. The homepage does not name programming languages or SDK packages. Confirm supported SDKs and API operations in Momentic’s docs before designing a direct integration. Its published integration surfaces are the CLI, MCP server, API keys, YAML test files, and CI workflow; the product also supports execution on local machines and hosted infrastructure. This gives teams a repo-centered path, but teams building custom tooling should verify API details and language support first.

Free costs $0 per month and includes $50 monthly usage, 30-day results retention, five concurrent browsers, and two Android plus two iOS sessions. Scale costs $500 per month, billed in advance, and includes $500 monthly usage, 25 concurrent browsers, five Android and five iOS sessions, and five phone numbers for SMS/OTP. Enterprise uses custom test-based pricing and adds SAML SSO, SCIM, an audit log, higher API rate limits, priority support, an SLA, and onboarding.

Usage can draw down the included balance: test steps cost $0.0125 each, AI-generated or failure-recovery steps cost $0.025, and hosted browsers cost $0.005 per minute. Large/WebGL browsers cost $0.0067 per minute; Android emulators cost $0.10 per minute and iOS simulators $0.1875 per minute. Scale top-ups cost $50–$1,000 each, with a $1,000 monthly limit; funds do not expire. Runs pause at a zero balance, with no overage billing. The pricing page lists AI test selection at $0.10 per call, result classification at $0.39, and triage/auto-healing at $1.25.

Momentic’s public homepage does not publish a source-code licence or say that the platform is open source. It describes hosted infrastructure and running tests on your own machines or in CI, but that is not the same as self-hosting the Momentic service. Ask the vendor whether a self-hosted control plane is available and what deployment options Enterprise includes.

The homepage also does not describe data export or a migration path. YAML tests live in your repo and can be submitted as pull requests, which makes the test definitions accessible outside the platform. But the agent, execution service, and any proprietary test features may still create migration work. Before adopting it for a large suite, confirm how results, recordings, and test metadata can be exported, and whether tests remain runnable without Momentic.

The published plan limits include five concurrent browsers on Free and 25 on Scale. Mobile session counts are two Android plus two iOS on Free, and five of each on Scale. Enterprise lists one-hour mobile sessions and higher API rate limits, but the vendor does not publish numerical API limits in the pricing details. Check concurrency, session duration, and queue behavior against peak CI demand before making runs required checks.

The vendor lists SAML SSO, SCIM, audit logs, priority support, and an SLA for Enterprise; confirm the exact terms and compliance coverage with sales. Runs pause when the usage balance reaches zero, so a depleted balance can stop test execution rather than trigger overage billing. Reddit comments are mixed: one commenter said their team chose Playwright with Copilot instead, while another questioned whether automatic updates would stay organized long term. Treat maintenance behavior and failure recovery as pilot criteria.