Zoho unveils Zia LLM to power smarter artificial intelligence for businesses

Zoho launches Zia LLM, a business-focused large language model powered by NVIDIA, boosting privacy and hands-on artificial intelligence for Indian enterprises.

Zoho has introduced Zia LLM, its proprietary large language model designed specifically for business applications, marking a significant step forward in enterprise artificial intelligence for the Indian SaaS giant. Announced at Zoholics India, Zia LLM is built entirely on NVIDIA´s artificial intelligence infrastructure and aims to address a spectrum of operational needs — from summarizing complex data and generating code to smart information retrieval.

The Zia LLM family features three distinct models, scaling up to 7 billion parameters. This modular approach lets businesses select the right balance among speed, cost, and computational performance to suit their requirements. Unlike conventional cloud-based artificial intelligence offerings, Zia LLM is deployed on Zoho´s own servers. This architecture is designed to ensure enterprise customers benefit from enhanced data privacy and control, a key concern in the rapidly evolving artificial intelligence market where sensitive business information is at stake.

Alongside the core large language model, Zoho has bolstered its suite with several complementary upgrades. Notably, there is a new Hindi-English speech recognition tool to better support India´s multilingual landscape. Businesses also get no-code artificial intelligence agent development tools, expanding accessibility for users lacking technical expertise. Over 25 pre-built Zia agents come bundled, enabling rapid deployment of intelligence across workflows. These combined features underscore Zoho’s ambition to deliver practical, privacy-respecting artificial intelligence that´s locally controlled and readily customizable for Indian and global enterprises.

59

Impact Score

Microsoft adds FIDES security to Agent Framework

Microsoft has released FIDES in Agent Framework to block prompt injection and data exfiltration with deterministic policy enforcement. The feature labels content by trust and confidentiality, then checks tool calls before sensitive actions can run.

Pope Leo XIV to publish encyclical on Artificial Intelligence

Pope Leo XIV’s first encyclical, “Magnifica Humanitas,” is set for release May 25 and will focus on Artificial Intelligence and the protection of human dignity. The Vatican will mark the publication with an unusual press conference featuring the pope, senior cardinals, theologians and an Anthropic co-founder.

AMD starts Venice production on TSMC 2 nm

AMD says its next-generation EPYC processor, Venice, is ramping production in Taiwan on TSMC’s 2 nm process technology. The company also plans a future production ramp at TSMC’s Arizona fabrication facility for data center and Artificial Intelligence infrastructure.

Tech researchers challenge Trump visa policy over online safety work

A lawsuit from the Coalition for Independent Technology Research is challenging a Trump administration visa policy that critics say targets fact-checking, trust and safety, and disinformation research. The case could shape how researchers, platforms, and the public understand online harms and free speech.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.