At a glance
Functionize prints credit counts per plan at functionize.com/pricing; its integrations page and MCP waitlist show which connections are live today, since the docs site was not publicly reachable.
- Approach
- AI agent with vision, no selectors
- Writing a test
- Plain-English steps, CLI, MCP, or dashboard
- Native mobile
- iOS simulators + Android emulators
- Public pricing
- Free tier, then $99/mo self-serve
- Pull requests
- GitHub App: run per PR, check + comment
- Approach
- Agent observes the live app and heals selectors
- Writing a test
- Describe intent in Studio or via coding agent skill
- Native mobile
- Web UI workflows; native mobile not itemized
- Public pricing
- Free $0, Pro $20/mo, Team from $40/user/mo
- Pull requests
- PR testing claimed; integrations listed coming soon
What a run returns
Every TesterArmy run ends in one of three verdicts, with the video, screenshots, and per-step transcript attached.
- Passed
Every step completed and every assertion held.
- Failed
A step or assertion did not hold, and the report shows which one.
- Blocked
The agent could not reach a verdict because the environment, credentials, or seed data were missing; blocked runs stay out of failure counts.
Where they differ
Both put an agent in front of your running app. They differ on what a unit of usage buys, which integrations are live, and whether a selector layer sits underneath.
The unit you pay for
Functionize credits pay for authoring, execution, and healing at an unpublished rate. TesterArmy counts one run per test, 250 on Hobby and 1,000 on Startup.
MCP and CLI availability
Every Functionize plan lists API and MCP access, but the MCP page is a waitlist. TesterArmy's MCP server, CLI, and REST API are live now.
Pull request checks
Functionize says every PR is tested, but its integrations page reads Coming Soon. TesterArmy's GitHub App leaves each PR a check and a video comment.
Selectors under the agent
Functionize Studio re-observes the app when the UI changes and repairs the selector. TesterArmy's agent keeps no selectors and reads the screen on every run.
Side by side
Functionize wraps Studio in an enterprise platform with a SOC 2 Type II report, SSO and RBAC, private deployment, and solution pages for Salesforce, ServiceNow, Workday, and SAP.
- How tests are writtenTesterArmyYou write plain-English steps in the dashboard, CLI, MCP server, or your coding agent.FunctionizeDescribe intent in Studio or through the coding-agent skill; the older Architect recorder remains.
- How the agent finds elementsTesterArmyThe agent reads the rendered screen like a user and needs no selectors.FunctionizeStudio captures DOM, state, and behavior per step, then repairs selectors after UI shifts.
- Pull request testingTesterArmyThe GitHub App tests each PR and deployment and posts a check with video.AdvantageFunctionizeThe developers page claims every PR is tested; the integrations page says Coming Soon.
- CLI, API, and MCPTesterArmyThe CLI (npm testerarmy), OAuth MCP server, and OpenAPI REST API all work today.AdvantageFunctionizeEvery plan lists API and MCP access, though the MCP page is a waitlist.
- Native mobileTesterArmyiOS simulators and Android emulators take .app and .apk builds, including Flutter and Expo.AdvantageFunctionizeFunctionize describes Studio as a web UI agent and does not itemize native mobile.
- Enterprise complianceTesterArmyTesterArmy has no third-party attestation yet, and SSO/SAML arrives with the Enterprise plan.FunctionizeFunctionize offers SOC 2 Type II, SSO and RBAC, audit logs, and private deployment.Advantage
- Reference customersTesterArmyThe customers page lists early-stage and growth engineering teams.FunctionizeNamed customers include Conduent, ServiceNow, McAfee, and Norstella.Advantage
- PricingTesterArmyThe free tier is followed by Hobby ($99/mo, 250 runs) and Startup ($299/mo, 1,000).FunctionizeIndividual $0, $20, or $100 a month; Team from $40 per user a month.
Facts checked September 2026. Spotted something out of date? Tell us and it gets fixed.
Pricing
Functionize sells an enterprise QA platform from a $0 plan, and it spends credits where TesterArmy counts runs. Both publish prices, so compare what one unit buys.
$99/mo after a free tier
- Free tier: sign up and run against your own app
- Hobby $99/mo: up to 250 test runs included, web and mobile
- Startup $299/mo: up to 1,000 test runs included, 10 concurrent test runs
Free $0, then $20/mo, credit-metered
- Individual plans run Free $0 (200 credits/mo), Pro $20/mo (400), Max $100/mo (2,000)
- Team plans cost Growth $40/user/mo (400 credits) or Scale $200/user/mo (2,000), with SSO and RBAC
- Enterprise plans carry custom credits, invoice or PO billing, and a 12-month minimum
Who should pick which
Both agents start free against your own app, so pick by what you need around the agent and how you want usage counted.
- Pick TesterArmy ifYou ship a SaaS or mobile app and want a run, check, and comment on every PR.Pick Functionize ifYou test Salesforce, ServiceNow, Workday, or SAP screens and want a vendor with pages and libraries for them.
- Pick TesterArmy ifYour coding agent (Claude Code, Cursor, Codex) should create and run tests through a CLI or MCP server.Pick Functionize ifYou need SOC 2 Type II, SSO and RBAC, audit logs, or a private deployment in your environment.
- Pick TesterArmy ifYou want to know before the month starts how many tests your plan covers.Pick Functionize ifYou want the agent to author tests from a description and keep them green without touching selectors.
Questions
Try a plan upgrade without spending a credit
Functionize counts credits on every agent action, with PR and MCP integrations still on the way. TesterArmy counts runs, and its MCP server, CLI, and GitHub App already work.