Rhesis AI
Verified Oct 15, 2026 ActiveOpen-source workspace where domain experts and engineers build test sets for AI agents together.
Listed on EffectHub ·
Pricing
- Pricing model
- Freemium
The Community edition is free and open source (MIT); Enterprise pricing (SSO, RBAC, API clients) is not published.
Use tool ↗About Rhesis AI
Rhesis AI is a testing platform for LLM and agent applications that brings non-technical domain experts into the same workspace as engineers, so both can write test scenarios instead of leaving coverage entirely to developers. It generates test cases including adversarial and edge-case prompts, runs them against an application, and traces failures back to a root cause. A red-teaming component called Polyphemus probes for jailbreaks, prompt injection, and PII leakage. The core platform is open source under MIT and self-hostable; an Enterprise tier adds SSO, role-based access, and API clients, though its price is not published.
- Platforms
- Web · API
- API
- Yes
- Last verified
- Oct 15, 2026
Verified sources
- Rhesis AI is a collaboration layer for domain experts and engineers to test AI agents, open source and self-hostablerhesis.ai ↗
- Core platform is MIT-licensed with a public GitHub repositorygithub.com ↗
Is Rhesis AI legit?
We don’t label tools safe or unsafe. Below are the records we checked ourselves, each with the date we read it.
- Domain registered
- Jan 4, 2024 (2 years ago)Registry record (RDAP), registrar Porkbun LLC, looked up Sep 14, 2026
- TLS certificate
- Issued by Google Trust Services, valid until Dec 6, 2026Seen in our own connection to the site on Oct 15, 2026
- Last checked
- Oct 15, 2026The site answered
- Operated by
- Rhesis AI GmbH, GermanyOur research, card verified Sep 1, 2026
Pros & cons of Rhesis AI
Rhesis is worth a look for teams that want domain experts, not just engineers, writing test cases for an AI agent, and who are comfortable self-hosting an open-source tool. The MIT-licensed core is genuinely free, but anything past the community tier requires talking to sales since Enterprise pricing isn't posted.
Pros
- MIT-licensed and self-hostable, not just a hosted demo
- Supports UI, SDK, and MCP access for different kinds of users
- Polyphemus red-teaming agent checks for jailbreaks and PII leaks specifically
- Integrates with common agent frameworks via OpenTelemetry instrumentors
Cons
- Enterprise pricing is not published anywhere
- Self-hosting the open-source edition requires your own infrastructure and setup time
Best for
- Teams that want product or domain experts writing test scenarios directly
- Engineering teams already using OpenTelemetry-based tracing
- Anyone needing a self-hostable, open-source eval platform
Not for
- Teams wanting a fully managed SaaS with published pricing
- Solo developers who just need a quick prompt-testing script