AI Sandbox Pilot Project
Service Overview
The AI Sandbox Pilot Project is an ecosystem (Sandbox) of AI tools and infrastructure open to all OU faculty, staff, researchers, and students. Launched in Fall 2025 by the Digital Scholarship and Data Services (DSDS) team, this pilot provides access to several AI testing environments, including no-code interfaces, API access, and GPU workstations. Workshops are also available to offer hands-on training for implementing AI workflows in research and projects.
What We Offer
The AI Sandbox supports both no-code and API-based platforms, enabling you to learn, experiment with, and integrate AI into your research, teaching, and creative workflows. You'll have hands-on access to generative and non-generative AI tools—including LibreChat, LiteLLM, Hugging Face, AWS Bedrock, and the National Research Platform (NRP)—all supported by specialized training materials and workshops.
- Artificial Intelligence Foundational and Carpentries Workshops
- LibreChat and LiteLLM Applications (Pilot)
- Lab with Three High-Performance AI Workstations (Pilot)
- Accessing Hugging Face models on NRP-hosted Nautilus platform
Get Started
Who Can Use This Service
The AI Sandbox is open to all current OU faculty, staff, researchers, and students for research, development, and learning purposes. For any additional information, please contact the DSDS team.
How to Get Access
To access the AI Sandbox applications or the AI workstation lab, submit the AI Resources Request Form to tell us which ones you’d like to use. We will follow up to provide you with access.
What's Included
The AI Sandbox features five integrated services spanning the full AI workflow — from exploration to production research.
LibreChat
No-Code ChatGPT-Style Interface
LibreChat provides an accessible chat interface to interact with both open-source and commercial large language models. It’s designed for researchers, instructors, and students who want to explore AI capabilities without writing code.
- ChatGPT-style UI
- Model switching between National Research Platform (NRP) and AWS Bedrock models
- Multimodal support (text and/or image inputs)
- Persistent session history and chat export
- Secure OU login via the university single sign-on (SSO)
Note: OU users can sign up with their 4x4.
LibreChat Terms of Service & Privacy Policy
Use Cases: Brainstorming, everyday use, drafting, research ideation, multimodal interaction, workshops, and classroom instruction.
LiteLLM
API Access for Advanced Users
LiteLLM provides OpenAI-compatible API access to multiple AI models, enabling programmatic workflows for research, development, and automation. To get access, contact the DSDS team or submit a Resource Request.
- Access to AWS Bedrock and NRP-hosted models
- Users can create virtual API keys
- Admin cost limits and usage tracking
- Multimodal support (text, images, file URLs, code snippets)
- Platform agnostic
Note: Please select “Request access to LiteLLM application” option, and we will contact you with an invitation link.
Hugging Face + Nautilus
Open-Source AI at Scale
Leveraging OU Libraries’ Nautilus Platform and Hugging Face, this service provides scalable, containerized environments for training and running open-source AI models.
- JupyterLab with PyTorch, TensorFlow, and Hugging Face (HF)
- GPU-enabled workflows through the National Research Platform (NRP)
- Exploratory and production research support
- Use Hugging Face ‘Tasks’ – e.g., Audio, Vision, or Text-related tasks
- Automate custom workflows with HF models
Log In To OU Nautilus Researcher
Nautilus + Hugging Face User Guide
AI Workstation Lab
Local Research Infrastructure
The AI Workstation Lab features three high-performance workstations (Mac Studio and Dell Pro Tower) equipped with advanced GPUs, available for AI research and education. To use the lab, you must accept the AI Workstation Lab Terms of Service, complete the AI Resources Request Onboarding Form (please select “Request to access AI Workstation Lab – Onboarding” option), and attend a brief consultation with a DSDS team member. After submitting the Onboarding Form, a team member will contact you to schedule the appointment.
- NVIDIA RTX 5000 ADA GPU (Dell Pro Tower)
- 2 × M3 Ultra Mac Studio with 60-core GPU
Note: Please select “Request to access AI Workstation Lab – Onboarding” option to get started.
AI Workstation Lab Terms of Service Policy
Already On-boarded?
Book a Workstation
Note: Access to the lab and booking privileges are limited to users who have completed onboarding. Approved users may reserve the lab at any available time.
AI Workshops and Carpentries
Foundation and Advanced Training
We offer both foundational and advanced workshops to help the OU community effectively use the AI Sandbox technologies. Workshops are built on existing Carpentries programs and customized AI sessions.
- Intro to Unix Shell & Python
- Programming with LLMs (APIs and Python)
- Docker & Containers for AI
- Responsible AI and Academic Integrity
Contact Information
Digital Scholarship and Data Services Team
FAQ
Request Access
AI Sandbox Applications Access
To access the AI Sandbox applications, submit the AI Resources Request Form to tell us which ones you’d like to use. We will follow up to provide you with access.
AI Workstation Lab Access
If you'd like access to the AI Workstation Lab, you must:
- Read and agree to the AI Workstation Lab Terms of Service Agreement by completing the form at the bottom.
- Submit the AI Resources Request Onboarding Form. (Please select “Request to access the AI Workstation Lab – Onboarding”)
- Attend a brief consultation with a DSDS team member. After submitting the Onboarding Form, a team member will contact you to schedule the appointment.
- After gaining access, you can reserve an AI workstation in the lab.
Review User Guides
If you're just getting started and want to learn more, please visit our user guides.
All current OU faculty, staff, researchers, and students are eligible for research and learning purposes. For any other cases, email the DSDS team.
NRP-hosted AI models are open source and available at no cost. OU Libraries funds the hosting of the AI Sandbox environments and covers the usage of AI models provided through AWS Bedrock.
To get help, use the following links: