Home

OpenVals Wiki

Welcome to the OpenVals documentation.

OpenVals is an open-source framework for evaluating, validating, benchmarking, and securing AI models.

This wiki is designed to help you:

• Understand how OpenVals works
• Set up and run evaluations
• Benchmark multiple models
• Extend the framework for your needs

---

Getting Started

• Installation
• Quick Start

---

Benchmarking

Benchmarking allows comparison across multiple models.

---

Features

• Run multiple models on same dataset
• Normalize results
• Rank models

---

Output

• Raw scores
• Normalized scores
• Final ranking

Core Concepts

• Evaluation Engine
• Benchmarking
• Metrics
• Scoring and Ranking

---

Advanced

• Custom Models
• Custom Metrics
• Dataset Format

---

Project Vision

• Why OpenVals
• Roadmap

---

Contributing

Contributions are welcome and appreciated.

---

Documentation

• [Look at doc.openvalidations.com]

More resources:

DrPinancle

CyberMind

OpenVals