GPT-5.6 LunaLightweight and fast; best for chat and high-volume work.
✓
✓
GPT-5.6 Sol
—
✓
GPT-6 Sol
✓
✓
GPT-5.6 Terra
✓
✓
GPT-5.5Best for advanced agentic work and planning.
✓
✓
GPT-5.4Best suited for heavier use, professional work, or longer tasks.
✓
✓
GPT-5.4 MiniFast; best for efficient general use.
✓
✓
✓ Available · — Not available. Usage is metered monthly; higher-cost models use your allocation faster.
Announcements
HokieAI Usage Limits Are Now Combined
New
HokieAI has changed how usage limits work. Previously, API usage and ONEchat each had a separate limit. Those limits are now combined into one shared limit that covers both ways of using HokieAI.
This means your API activity and your ONEchat activity now count toward the same usage allowance. Rather than thinking about two independent limits, consider your total usage across both. If you use HokieAI through the API and work in ONEchat, both contribute to the combined limit.
As you plan your HokieAI use, keep both kinds of activity in mind. A workflow that includes API requests alongside conversations in ONEchat should be considered together, rather than as two separate pools of usage. This is especially important when deciding whether the shared limit will meet your needs.
If you need a higher combined limit, you can request an increase through 4Help. Visit the request form to submit your request.
The takeaway: API usage and ONEchat now share one HokieAI limit. Please account for both when planning your work and use the request form if you need additional capacity.
Announcing the new HokieAI Community, a public Microsoft Teams team where you can connect with fellow HokieAI users, ask questions, and work together to find answers. Whether you are exploring an idea or working through an implementation challenge, this community offers a place to learn through conversation and collaboration.
We encourage you to join and collaborate by sharing your experiences, asking questions, and helping one another. Bring the approaches you have tried, the challenges you have encountered, and the insights you have gained. Your experience may help someone else move forward, while another member’s perspective may give you a new way to approach your own work.
Many implementation approaches, practical examples, and lessons learned are best discovered through collaboration with fellow users. Sharing what worked and what did not can spark useful discussions and help members consider different options. You do not need a finished solution to participate; a thoughtful question or an example from your experience can be a valuable contribution.
Join the HokieAI Community and be part of the conversation. Exchange ideas, learn from others, and contribute what you know.
Together, we can build shared understanding and make AI exploration a more collaborative experience.
API token budgets are now managed the same way as ONEChat token budgets.
Requesting token increases is simpler for everyone, and all budget usage (API and ONEChat) appears in the Monthly Usage limit icon in the top-right corner.
Redesigned Agent Builder
Agent details, conversation starters, and other configuration options are now managed in the Agent Builder.
Improved chat experience
Users can view the date and time for individual chat messages.
Time-period headings remain visible while scrolling through chat history.
Improved keyboard accessibility
Users can navigate the chat message toolbar with the Tab, arrow, Home, and End keys.
Important Fixes
Token usage is now credited to the model that actually processed the request.
Users can type and submit a new prompt while a response is generating.
Stopping a response now preserves the prompt and partial response in the conversation.
Large text files now provide a warning when content must be shortened.
GPT-5.6 Luna is now available in HokieAI alongside GPT-5.4 mini, GPT-5.4 and GPT-5.5. It is the lowest-cost option and well suited to chat and high-volume work. Additional models will be added as more options are evaluated.