Dear,
Not sure which LLM model is being used by OAC AI agents, would it be possible to have an option to select a LLM model for all/some agents?
Regards,
Bhavin Desai
Integration with other LLMs would be interesting to see
In the console you can select from supported models and as of the July release you can also register an OCI dedicated cluster running GPT-OSS and use that for OAC AI features. However, the user cannot dynamically pick a model at runtime like many familiar tools. Also, we only allow models that we have certified as we want to assure a level of quality that not all models can achieve.
Right now once you save a workbook, that's it. If you (or someone else with edit access) changes a canvas, deletes a calc or messes up a filter and saves, there's no easy way to get back to how it was. The options are digging out an old .dva export if you happened to take one, or asking an admin about a snapshot, which…
There are more menu options made readily available by default to non-admin users that we wish to restrict. It seems this functionality, to control menu options in the user button menu on the upper-right-hand-side of OAC (DV), is not currently available. Can we be given the ability to customize this menu? Ideally via…
Right now, OAC offers pre-built pages and watchlists related to NSAW. At our University, we do not need these pages, and we want to delete them entirely. We've been told in response to our SR that all we can do is keep "Access" limited to only the default admin and sysAdmin roles. This prevents regular users from seeing…
It would be nice to have a way to right click on an object (subject Area, Column, Calculation, etc.) in the Data Panel and show where that object is referenced in a workbook = canvas, visualization, filter, calculation, data action, etc.) In Workbooks with multiple canvases and multiple subject areas, we are left to hover…
When Oracle Analytics AI features find an anomaly, trend or key driver, let authors save those results as a dataset or subject area. Each run would add rows with the metric, the date found, the expected versus actual value, the severity, the main drivers and the filters applied. Authors could then build workbooks, schedule…