Personal AI at NVIDIA GTC: Bringing Memory to the AI Grid

A Milestone Moment at GTC

NVIDIA GTC 2026 was a defining moment for Personal AI. We announced a strategic partnership with Comcast and NVIDIA to advance the AI Grid — a validated framework that leverages distributed network infrastructure to accelerate next-generation AI delivery. In this new era, tokens are replacing data and agents are replacing applications. For Personal AI, this means bringing memory-based Small Language Models to the network edge for real-time, hyper-personalization that is private and economical with unmatched reliability and performance.

Benchmarking at a Glance

Benchmarks from Comcast's AI Grid trials using Personal AI's voice SLM on NVIDIA RTX PRO 6000 GPUs validate the approach: 

• Sub-500ms end-to-end latency, maintained even at P99 burst traffic
• Up to 40x lower operating costs vs. centralized LLMs
• 27x faster time to first token (TTFT)
• Significantly higher throughput, generating 42K tokens per second 

To get up to speed on what was announced with Personal AI at GTC, check out our handy list of relevant resources.

>> Learn more about our collaboration with Comcast and NVIDIA:

• Personal AI Press Release:
  Personal AI’s Memory-Based Small Language Models Deliver Hyper-Personalized Experiences on Comcast’s AI Grid, Powered by NVIDIA

• Comcast Press Release:
  ‍Comcast to Accelerate Next-Generation AI Applications Using NVIDIA Infrastructure at the Network Edge
  ‍
• HPE Press Release:
  ‍HPE transforms distributed AI factories into intelligent AI grid powered by NVIDIA

>> Take a deep dive into our benchmarking findings and explore the impact of our work with Comcast and NVIDIA on the future of AI:

• Personal AI Blog Post:
  ‍Personal AI’s Memory Platform: Monetizing The NVIDIA AI Grid

• NVIDIA Corporate Blog Post:
  ‍NVIDIA, Telecom Leaders Build AI Grids to Optimize Inference on Distributed Networks

• NVIDIA Technical Blog Post:
  ‍Building the AI Grid with NVIDIA: Orchestrating Intelligence Everywhere

>> Watch key session recordings through NVIDIA GTC On-Demand (login required)

• NVIDIA Telecom Keynote (Ronnie Vasishta, SVP of Telecom)
• Comcast AI Grid Session (Nader Foroughi, Distinguished Engineer)
• Personal AI Distributed Inference Session (Sharon Zhang, Co-Founder & CTO)

>> Hot off the presses, see what’s been covered in the news:

• Our Press Page