Release notes and product updates
Latest Gemini release v7.6.1 — October 2025
Gemini now lets you configure Entra Id Authentication. When you do this, you will no longer be challenged to login to Gemini directly, instead you will be routed to login.microsoftonline.com where Azure’s Entra logic will take over. If you are successfully authenticated, Microsoft will pass Gemini your credentials and those credentials will be used to look you up and log you in. It is SSO in the same way that Windows Authentication is, only the authentication is with Azure and not with IIS.
Gemini now offers the option of configuring your SMTP and Mailboxes to use the Graph API in addition to IMAP, EWS, and POP. It is simpler to configure, faster to validate, and it comes out in time for our customers not to be impacted by Microsoft announcing the deprecation of EWS in October 2026.
For those customers who use free public SMTP services, such as smtp.gmail.com and smtp.office365.com, we have introduced message throttling. These services impose limits on the number of messages that can be sent per minute, and Gemini allows you to regulate the maximum number of emails it will send in a minute.
There was a recent issue with Microsoft OAuth tokens being truncated because Microsoft changed the format of the tokens (which is their absolute right), and the encryption meant the length exceeded what the application expected to store. This has been fixed in 7.6.1
We addressed two security vulnerabilities in versions up to and including Gemini 7.5, one of which is quite obscure, requires a high degree of expertise to exploit, and would elude many (if not all) Penetration Testing tools. We are grateful to Robert C. Raducioiu from 7Layers S.r.l. for his diligent research and testing.
Note: For the protection of all our customers, it is our policy not to describe vulnerabilities. Not all customers will upgrade to the latest release, and our on-premise customers will never all upgrade at the same time.
Under the hood and behind the curtain, Gemini is expanding...
Lots of customers have enquired about integrating Gemini data into AI models. We are currently researching the use of SQL Server’s Vector data types and/or JSON arrays to store data in a format that allows similarity comparison queries. This is more complex than it might first appear, as there is usually a huge legacy data volume that should be ignored, and there are questions about security to be resolved – if a user can’t see a ticket, should they be told that the answer to their query is in a comment on it? Our goal is to provide optional solutions to such problems in a manner that our customers control.
Upgrade to the latest Gemini product release.
Cloud pricing starts from just $10 per user per month.
Self-host perpetual pricing starts from just $850 for 10 users.
