Magnitude
By Magnitude (magnitudedev). Category: Agentic end-to-end testing. Homepage
The magnitudedev/browser-agent repository describes Magnitude as an open-source, vision-based browser agent with a test runner for web apps. It is licensed under Apache-2.0. Test cases are written as natural-language steps in .mag.ts files. The magnitude-test runner executes them with Playwright and calls an LLM at run time. Repository docs cover running tests in GitHub Actions. A magnitude-mcp package exposes browser actions to MCP agents such as Claude Code and Cline. The Magnitude name and website now describe a model inference engine. The testing repository's last releases are dated 2026-02-08.
Description written from the vendor's documentation.
Status
Status: Active
The Magnitude name, the magnitudedev/magnitude repository and magnitude.run (which redirects to magnitude.dev) now describe an open-source local model inference engine. The browser-agent and test runner code lives in magnitudedev/browser-agent (not archived); its last releases (magnitude-test 0.3.13, magnitude-core 0.3.1, magnitude-mcp 0.1.3) are dated 2026-02-08, and the docs domain docs.magnitude.run no longer resolves.
Facts
| Fact | Value | Evidence | Source | Verified on |
|---|---|---|---|---|
| Status | Active | Documented | github.com | |
| AI capabilities | Documented | github.com | ||
| Test creation | Documented | github.com | ||
| Test output | Proprietary runner | Documented | github.com | |
| Surfaces | Documented | github.com | ||
| Hosting | Documented | github.com | ||
| Pricing model | Documented | github.com | ||
| Free tier | Yes | Documented | github.com | |
| CI integrations | Documented | github.com | ||
| Self-healing | None documented | Not documented | None found | |
| Agent integration | magnitude-mcp: MCP server exposing Magnitude's vision-based browser actions to MCP-compatible agents (Claude Code, Cline and others) | Documented | github.com | |
| Open source | Yes, licence Apache-2.0 | Documented | github.com |
Published prices
No prices are published.
Vendor-stated, verbatim
What Magnitude (magnitudedev) says about its product, quoted as published and not confirmed in documentation.
“Magnitude is state-of-the-art, scoring 94% on WebVoyager!”
Alternatives
Tools in the same category (Agentic end-to-end testing), ordered by the number of shared creation and surface values, then by name. Each line lists documented facts that differ from Magnitude.
- ACCELQ: documents creation by recording; hosting cloud, hybrid
- AISquare Studio QA: documents output Standard Playwright; creation by code generation
- Autify: documents output Mixed; creation by prompt to test, recording; hosting cloud
- Autify Nexus: documents output Mixed; creation by recording, prompt to test; hosting cloud
- Autonoma: documents hosting cloud
- BlinqIO AI Test Engineer: Agentic end-to-end testing; no documented difference in output, creation, hosting or free tier
Questions
Is this page a review of Magnitude?
No. This page is a fact sheet. Each fact shows its evidence level, its source and the date it was checked. We do not rate or rank tools.
Does Magnitude have a free tier?
Yes. Magnitude has a free tier. This is documented, checked on 2026-09-24.
What test output does Magnitude produce?
Output: Proprietary runner. This is documented, checked on 2026-09-24.
Is Magnitude still available?
Status: Active, checked on 2026-09-24.
Sources
- magnitudedev/browser-agent README (GitHub) Opened
- magnitudedev/browser-agent releases Opened
- magnitudedev/browser-agent LICENSE Opened
- Building Test Cases (docs source in repo) Opened
- Testing in CI (docs source in repo) Opened
- Magnitude MCP README Opened
- magnitudedev/magnitude README (GitHub) Opened
- Magnitude homepage (magnitude.run redirects here) Opened
- magnitude-test on npm Opened
Spotted an outdated or wrong fact about Magnitude? Send the source and we apply it at the next monthly update.