The University of Florida provides access to several AI applications to its students, faculty, and staff. Click on the tabs above to learn more about each of the following:
The NaviGator AI suite provides self-service access to AI services for UF students, faculty and staff.
NaviGator ChatNaviGator Chat is a web application that provides UF students, faculty, and staff the ability to chat with several large language models (LLM) and use image generation models with their own data sets.
NaviGator ToolkitNaviGator Toolkit is a collection of AI tools that help UF students, faculty, and staff kickstart their custom applications to make them AI-powered. Additionally, Toolkit provides UF students, faculty, and staff the ability to use several large language models (LLM), image generation models, and text to speech/speech to text models via the UF AI Gateway.
Microsoft's Copilot is built on OpenAI's ChatGPT models.
UF Microsoft Copilot Microsoft Copilot combines the power of large language models (LLMs) with up-to-date information from the web.
The Primo Research Assistant uses content found in the Smathers Libraries' collection to identify 5 documents that can help answer your question.
Primo Research Assistant
Dimensions offers multiple AI tools, including a custom instance of ChatGPT
Dimensions Click "Login" at the top of the page, Enter your UFL e-mail address and next. You will be prompted to login with your GatorLink username/password.
NaviGator Chat allows you to use different Large Language Models (LLMs) with your own dataset. You can use AI to discover trends and patterns, look for insights, and produce reports based on your data.
UF provides this service to allow you to analyze your documents using different language models while keeping those documents secure within UF servers and contracted vendors. At this time, UF permits the usage of restricted or sensitive data with models deployed on UF HiPerGator. The usage of restricted or sensitive data is not permitted with cloud models. When interacting with the language models hosted by vendors, your messages and subsets of your documents will be sent to a LLM instance provided by Microsoft or Google. All data handled in this fashion is covered by our existing agreements with Microsoft and Google. None of this data contributes to training the large language model. No data is retained after deletion. If you have uploaded a file by accident, deleting the file or conversation will completely remove it from the system.
UFIT offers a set of AI tools to allow custom and SaaS applications to utilize large language models such as GPT, Llama, Gemini, and Claude. Additionally, image generation models like DALL-E and speech to text models like Whisper are also available.
NaviGator Toolkit provides every UF student, faculty, and staff the capability to generate API keys and leverage AI models in their current applications. Each UF user has access to locally deployed models such as Llama, Mixtral, Gemma, Codestral, Stable Diffusion, and Mistral. Additionally, researchers can be onboarded as teams with custom budgets to leverage cloud models. Researchers are encouraged to submit a request via the UFIT Help Portal.
UF provides this service to allow you to analyze your documents using different language models while keeping those documents secure within UF servers and contracted vendors. At this time, UF permits the usage of restricted or sensitive data with models deployed on UF HiPerGator. The usage of restricted or sensitive data is not permitted with cloud models. Users only have access to their own datasets and conversation history. When interacting with models hosted by vendors, your messages and subsets of your documents will be sent to a LLM instance provided by Microsoft, Amazon, or Google. All data handled in this fashion is covered by our existing agreements with Microsoft. Amazon, and Google. None of this data contributes to training the large language model.
This service currently provided by Microsoft Copilot (not M365 Copilot) is protected by UF-Microsoft data agreement. Copilot is an AI assistant that uses OpenAI's GPT-4 model. It generates human-like responses and functions as a web search engine with access to web search results and up-to-date information. To start using it, click the link below and log in with your GatorLink. Available to UF Students, Faculty, and Staff.
When used with your GatorLink login Microsoft provides additional protection of your data above and beyond the basic publicly available Microsoft Copilot service. At this time, UF does not permit the usage of restricted or sensitive data with UF GPT powered by Microsoft Copilot, please see the UF Data Guide for more information. Conversation history is not retained post each session, and user interactions do not contribute to training the large language model. Further work is underway to assess the service for use with additional data classifications. Check back here for further updates.
The Primo Research Assistant is a tool powered by Generative Artificial Intelligence (specifically, Large Language Models or LLM). It allows you explore academic content by asking questions in natural language. The tool uses most of the content found in your library to identify five documents that can help answer your question. It then extracts the most relevant information from the description/abstracts of each source to write the answer. Above the answer, you’ll see the sources used to generate it along with in-line citations that let you clearly see which source was used to generate each part in the answer. Use these sources to delve deeper into the topic and to fact check the responses from the tool. Keep in mind that AI-generated answers should be verified for correctness, as they may not always be accurate or up-to-date.
The Primo Research Assistant is not a replacement for human expertise but uses artificial intelligence to automate otherwise time-consuming tasks. We’ve designed the Primo Research Assistant make it easier to understand topics, their context, and resources published about it. Use the “view related results on your library search” button to find more documents relevant to your question. Click the AI-generated “related research questions” to explore topics similar to your question.
Your question is converted into a query that the search engine understands with the help of a Large Language Model (currently GPT 4o-mini). The search engine then identifies the most relevant documents in the index. It ranks them according to how well they can answer the question and, again with the help of the Large Language Model, creates an answer from the top 5 sources.
Due to the nature of Large Language Models, answers to the same question are not always the same. There may be more than one possible answer and different resources that are relevant. If you are not satisfied with your answers, use the “Try again” button.
To make the most of the Primo Research Assistant, it's essential to ask clear and detailed questions about academic or scientific topics. Be as specific as possible and phrase your query in the form of a question. Example queries can be found on the starting screen.
The Primo Research Assistant supports local language searches. Most material in our index is in English. If you ask a question in another language than English, the Assistant will search in both, your local language and English, and write the answer in the language of your question. Note that there is a dependency on the Large Language Model and language support may vary.
The Research Assistant allows you to specify resource types, and dates, for example if you are looking specifically for books or articles about a topic, or works from a certain date range. The Research Assistant will use this information to filter the results accordingly. For example, you can ask for sources to only consist of peer reviewed titles (e.g. “what peer review articles talk about the influence of plastic waste on marine life”). Available filters are Articles, Books, peer reviewed and a date range. In addition, if you ask in your query for example for “recent” or “the latest” material, the Research Assistant will filter the results by what was published in the last 5 years. While the Research Assistant will pick up those words directly from your query, you can also use the filter function on the lefthand side of the search boxes ( “What is your research question” box, “Ask your next research question”). The filter is sticky and will be applied to subsequent queries unless you delete it, either by changing the value in the filter dropdown, or by using the recycling bin on the right side of the screen.
Tip: If you want to rerun the same search as before - for example after you chose a filter - you can click into the “Ask your next research question” box and hit the “arrow up” key on your keyboard.
The Primo Research Assistant does not yet support follow-up questions. Each question stands by itself. For example, if you ask “What topics did Simone de Beauvoir write about”, you cannot follow up by asking “and what is the philosophy of her work” and expect the system to understand what you mean. At this time, you will have to include all relevant information in each question, e.g. “what is the philosophy of Simone de Beauvoir’s work”?
The Primo Research Assistant is a work in progress. We encourage you to share your thoughts with us about what can be improved. Please use the thumbs up and down buttons underneath the answer to provide feedback.
|
