Zoho CRM adds Zia email summaries and switching between multiple LLMs in bots
Zoho CRM's May 2026 release notes add Zia Emails Summary, which condenses past email exchanges on a record, and the ability to deploy and switch between multiple language models within Zia bot frameworks.


Image: Zoho CRM press materials, from the announcement
Zoho CRM's May 2026 release notes contain two Zia entries. "Zia Emails Summary: Instant context from past emails" adds per-record summaries of past email exchanges. "Switch between multiple LLMs instantly for tailored Zia experiences" lets organizations deploy several language models in bot frameworks and switch between them. Zoho positions the second change as giving admins model control per use case and giving users choice inside the bot. Both entries link to community posts.
The details
- Zia Emails Summary condenses all past email exchanges for a record.
- Multiple LLMs can be deployed and switched within bot frameworks.
- Admins can assign models to use cases; end users can choose a model within the bot.
- The multi-model option is listed for Professional, Enterprise, Ultimate, CRM Plus and Zoho One.
- The May 2026 notes also add the Reports module to the iOS app.
Context
Zoho's Q1 2026 CRM update on March 9, 2026 described support for more language models in Zia's Smart Prompt. In January 2026 Zoho let teams bring QuickML models into CRM records through Custom AI Studio, with outputs shown in a widget on record details. Zoho launched Zia Chat as an AI interface across its applications on September 23, 2026.
Not disclosed
- Which LLMs are available to deploy in bot frameworks.
- Whether third-party model usage carries extra cost.
- The exact release dates within May 2026.
- Vendor
Zoho CRM- Category
- CRM & Marketing Automation
- Announcement type
- AI launch
Report an error
Vendors and readers can flag a factual error in this story, or ask for an image or logo to be removed. Email Cora, the editor, with the link. Image and logo removals happen within one business day. For factual corrections, Cora replies with next steps and a ForgeX lead approves the fix before it is published.




