Arthur Engine

What it does

Arthur Engine is an open-source toolkit for evaluating AI models and agent applications. Teams can configure tracing, prompt management, evaluations and experiments to inspect behavior and compare changes during development.

The engine can be installed with Docker, with additional deployment options documented in its repository. It is separate from joining the managed Arthur platform. Evaluation coverage depends on the selected metrics and test data, and running the toolkit still requires infrastructure and any external model services used by the application.

What it can do

• Tracing
• Prompt management
• Evaluation metrics
• Experiments
• Deployment templates

Before you choose

• Infrastructure and model-provider costs are separate from the free toolkit.
• Automated metrics require interpretation against real application requirements.

Listed under

• Security
• AI and agent security

Get started

1. Review the repository’s installation requirements.
2. Deploy the engine and connect a test application.
3. Configure evaluations and inspect results across relevant cases.

The provider describes Arthur Engine as free and open source. Infrastructure, model services and the managed Arthur platform have separate costs and terms.

We couldn’t find published prices for Arthur Engine. See their pricing page ↗

Sources for this overview

Product pages, documentation, and pricing checked 30 Oct 2026. Each source supports part of the overview; availability and prices can change.

• Setting up the Arthur Evals Engine | Arthur ↗www.arthur.ai

Company website

arthur.ai: product, pricing and policy pages

Social accounts

Accounts the company links, and its posts on X

Human review

Checked by a person

knowledge_base

manual_review

Last checked 30 Oct 2026. Every field keeps its source and date; unknown means we couldn’t confirm it.