---
title: README
emoji: 🔥
colorFrom: green
colorTo: gray
sdk: static
pinned: false
---

## GroundedAI

### Mission
At GroundedAI, we are committed to fostering the responsible development of grounded generative AI (GAI) applications. Our mission is to push the boundaries of GAI while ensuring it remains grounded in factual information and ethical principles.

### Open Source Initiatives

#### Evaluation Models
As part of our mission, we have trained and released small language models tailored for evaluating key metrics like faithfulness, RAG relevance, toxicity, hallucination, and more. These specialized models enable cost-effective and customizable evaluation of GAI outputs by leveraging continued training on user data. Our open source models aim to provide efficient, high-performing alternatives to black box large language models currently used for evaluating GAI applications.

#### GroundedAI Python API Client
To foster a vibrant ecosystem, we have open sourced our GroundedAI Python API client, empowering developers to seamlessly integrate our services into their applications. Comprehensive documentation and sample code accompany our releases, lowering the barrier to adoption.

### Services

#### Consulting
We offer consulting services to organizations seeking to harness the power of GAI, helping them build cutting-edge applications tailored to their needs.

#### Evaluation API
Our robust evaluation API leverages statistical methods and language models as impartial judges, ensuring AI systems are thoroughly assessed for potential risks, biases, and adherence to ethical standards.

### Commitment to Responsible AI
At the core of our vision lies a commitment to responsible AI development, where transparency, reproducibility, and accessibility are paramount. We aim to be a driving force in shaping the future of generative AI, grounded in ethical principles and fueled by collaborative efforts with the global AI community.