---
title: "Vera Rubin NVL72"
type: entity
entity_type: model
url: https://frontiernewswire.com/models/vera-rubin-nvl72
updated: 2026-10-02T14:12:41.787Z
---

# Vera Rubin NVL72

Canonical: https://frontiernewswire.com/models/vera-rubin-nvl72

- **Type:** Model
- **Signals:** 4
- **First seen:** Sep 9, 2026
- **Last seen:** Oct 2, 2026
- **Also known as:** NVIDIA Vera Rubin NVL72

## Recent signals

- [Multi-Rack NVIDIA Vera Rubin NVL72 Bring-Up \| CoreWeave](https://frontiernewswire.com/s/multi-rack-nvidia-vera-rubin-nvl72-bring-up-coreweave-7bd011e4) (Oct 2, 2026)
- [First Vera Rubin NVL72 Customer Sees 4.8x Throughput \| CoreWeave](https://frontiernewswire.com/s/first-vera-rubin-nvl72-customer-sees-4-8x-throughput-coreweave-4a45dd25) (Oct 2, 2026)
- [Vera Rubin NVL72 inference tests show up to 7x better token throughput per MW vs. Blackwell on a 1.6T DeepSeek model, above Huang's 3x claim for 1T-3T LLMs (Bryan Shan/SemiAnalysis)](https://frontiernewswire.com/s/vera-rubin-nvl72-inference-tests-show-up-to-7x-better-token-throughput-df1e60a0) (Sep 14, 2026)
- [To Infinity and Beyond: ThunderKittens Now on NVIDIA Vera Rubin NVL72\!](https://frontiernewswire.com/s/to-infinity-and-beyond-thunderkittens-now-on-nvidia-vera-rubin-nvl72-7634ec79) (Sep 9, 2026)

## Related

- [Nvidia](https://frontiernewswire.com/companies/nvidia) (Company)
- [AI Chips](https://frontiernewswire.com/topics/ai-chips) (Topic)
- [CoreWeave](https://frontiernewswire.com/companies/coreweave) (Company)
- [DeepSeek](https://frontiernewswire.com/models/deepseek-model) (Model)
- [Blackwell](https://frontiernewswire.com/models/blackwell) (Model)
- [Cognition](https://frontiernewswire.com/companies/cognition) (Company)
- [SemiAnalysis](https://frontiernewswire.com/companies/semianalysis) (Company)
- [Jensen Huang](https://frontiernewswire.com/people/jensen-huang) (Person)

---

[All models](https://frontiernewswire.com/models) · [Latest Signal](https://frontiernewswire.com/)