Skip to main content

Nutanix Enterprise AI 2.5 Release

  • November 7, 2025
  • 0 replies
  • 232 views

mikebarmonde
Nutanix Employee

🤗🤖🤗🤖🤗 NAI 2.5 is Here! 🤗🤖🤗🤖🤗

 

📢What’s new in NAI 2.5?

 

NAI Key Features SummaryList

🏅Latest Innovations

• NAI Labs Sample Apps: Chat, and Talk-to-My-Data
• Dynamically scaled endpoints (Edit existing endpoint to upscale instances)
 

🤖New AI Compute Options

• MIG for GPUs support
• vGPU support
• RTX PRO 6000 GPU support for AHV

 

🔨Platform Enhancements

• AD/LDAP/SAML support
• New licensing now accepted, and the free-trial license is replaced with 60-day POC keys
• Logging and rsyslog support

 

📎Usability Features

• All charts default to 24h now vs. 15m
• Can now search on imported models and created endpoints
• Can now sort on endpoint name and status. Ie. sort active to the top in one click
• Removed 'instance count' on endpoints page to save space. It's on the endpoint summary screen
• Replace model type with model tags
 

✅New Pre-Validated Models

New models for reasoning, text, and object detection

Total list of all Hugging Face models, including new models for NAI 2.5

 

Total list of all NVIDIA models, including new models for NAI 2.5

 

🔮One More Thing - NAI Gateway Early Access

New in 2.5 for early-access is the NAI Gateway with early access with load balancing, rate limiting, and API controls (API-Only, non-supported)

Abstract stock image of the NAI Gateway

😅What now?
 

Nutanix customers can download a 60-day license to play with on the support portal that can run on any Kubernetes: https://portal.nutanix.com/page/downloads/list

Get started with installing NAI here: https://portal.nutanix.com/page/documents/details?targetId=Nutanix-Enterprise-AI:top-nai-install-t.html

 

🔎What is Nutanix Enterprise AI?

Nutanix Enterprise AI (NAI) is a centralized inferencing control plane that deploys and manages LLMs and their endpoints (OpenAI-compliant APIs) that’s simple, secure, and sovereign helping you standardize generative AI infrastructure across an organization, including delivering AI resources to your applications, developers, and data scientists. NAI deploys on any CNCF-certified Kubernetes, that can run at the edge, datacenter, public clouds, or even in an air-gapped environment.

This topic has been closed for replies.