요약
NVIDIA's Vera Rubin is finally here, delivering unimaginable amounts of AI performance versus Blackwell as the first systems power on. NVIDIA Vera Rubin Now Deployed & Pumps Out 800,000 Tokens/s In DeepSeek-R1, Crushing Blackwell At The Same Wattage Well, it'…
본문
AI Hardware Industry
NVIDIA Vera Rubin NVL72 Enters The Stage With A Monstrous 10x Uplift In Token Throughput Versus Blackwell, Achieves 800,000 Tokens/s Vs GB200’s 80,000 at The Same 150MW
Hassan Mujtaba •
Jul 21, 2026 at 11:00am EDT
Add Wccftech on Google
Comments
NVIDIA's Vera Rubin is finally here, delivering unimaginable amounts of AI performance versus Blackwell as the first systems power on.
NVIDIA Vera Rubin Now Deployed & Pumps Out 800,000 Tokens/s In DeepSeek-R1, Crushing Blackwell At The Same Wattage
Well, it's here: the most powerful AI platform that the world has seen so far, Vera Rubin. Part of NVIDIA's Extreme Co-Design initiative, which aims to deliver a full-fledged stack of AI-ready hardware and software, enabling unprecedented levels of performance per watt & delivering the lowest token cost, Vera Rubin ushers in a new era of AI.
Related Story NVIDIA Blackwell GB300 Continues To Set World Records for MoE Pre-Training While GB200 Sees A 4x Boost In Perf/W Through Continuous AI Software Stack Optimizations
The NVIDIA Vera Rubin platform is a diverse set of hardware and software solutions, which are the realization of the five-layer cake framework forming a comprehensive AI ecosystem. The stack houses a total of six trays, with some of the world's most advanced chips ever made:
- NVIDIA Vera Rubin NVL72 Compute Tray (Rubin + Vera + Grace Bluefield + ConnectX-9)
- NVIDIA Vera Rubin NVL72 NVLink Switch Tray (NVLink6 Switch)
- NVIDIA Vera CPU Tray (Vera)
- NVIDIA Spectrum-6 SPX Switch Tray (Spectrum-6 Switch)
- NVIDIA Groq 3 LPX Tray (Groq 3)
- NVIDIA Vera Bluefield-4 STX Storage Tray (Vera Bluefield)
NVIDIA's rack-scale solution process has also seen continuous refinements. Just like how the Raptor engines got refined by SpaceX, resulting in the Raptor 3, which is a sleeker and neater version with more advanced capabilities, NVIDIA refined its Vera Rubin trays, eliminating the use of cables, fans, and housings, in favor of a more streamlined and liquid-cooled system that fits into the rack seamlessly with a modular approach. This shows how the engineering efforts went beyond just chips and code.
Now, NVIDIA's first Vera Rubin NVL72s are powering on, and the first results are coming in. Last month, we reported how Coreweave and other major cloud providers were installing & validating NVIDIA's Vera Rubin platforms; well, today, we get to see how they perform too.
In the first DeepSeek R1 tests shared by Coreweave, the company has managed to sustain a 10x higher token throughput per watt versus Grace Blackwell. An NVIDIA GB200 NVL72 Grace Blackwell system managed just 80,000 Tokens/s at 150MW. The NVIDIA VR200 NVL72 Vera Rubin platform offers 800,000 tokens/s at the same wattage, showcasing a massive increase in MoE workloads.
And this is only the beginning; NVIDIA's Vera is all set to disrupt the market when more systems come online and showcase their potential. Backed by a strong suite of networking solutions that will enable companies to achieve Gigascale levels of AI, NVIDIA is once again showcasing why it is the leader in AI.
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
Further Reading
NVIDIA Rubin GPUs Bring 10x Increase in Agentic AI Performance Versus Blackwell as Its Architecture Gets Fully Unpacked, Featuring 336 billion Transistors
NVIDIA RTX Spark Gets First Developer Drivers as Native Windows on Arm Support Arrives Ahead of Fall Launch
NVIDIA GeForce RTX 50 SUPER GPUs Have Reportedly Arrived At AIBs, But Are On Hold Due To Undecided Memory Prices
U.S. Officials Reportedly Confirm First NVIDIA H200 AI GPU Shipments To China Two Months After Trump-Xi Meeting
Read all comments on NVIDIA Vera Rubin NVL72 Enters The Stage With A Monstrous 10x Uplift In Token Throughput Versus Blackwell, Achieves 800,000 Tokens/s Vs GB200’s 80,000 at The Same 150MWTrending Stories
-
AMD’s Zen 6 EPYC Venice Pictured as a 256-Core Monster, Massive Twin I/O Dies & First 2nm HPC Chip To Enter Volume Ramp
26 Active Readers -
PS5 Emulator SharpEmu Hits A Major Milestone, As Demon’s Souls Remake Reaches Character Creation
21 Active Readers -
Ubisoft Copies The Crimson Desert’s Playbook, As Assassin’s Creed Black Flag Resynced Ditches Roadmap For Community Feedback
19 Active Readers -
Sony’s PlayStation Vita Streams Halo and Forza After Its 2019 Demise, As Modders Turn It Into a Portable Xbox
18 Active Readers -
NVIDIA’s Jensen Quells Fears Surrounding Kimi K3 And Over-Investment In LLMs, Says “Everyone’s Got It Backwards, More AI Computing Power Will Be Needed”
16 Active Readers
Popular Discussions
-
AMD Medusa Point 10-Core “Zen 6” CPU Beats Strix Point 10-Core “Zen 5” By Nearly 35% While Operating at 5.4 GHz
2489 Comments -
AMD Ryzen 7 7700X3D 4.5 GHz “3D V-Cache” CPU Review: The Budget X3D Champ For AM5
1371 Comments -
AMD Unveils Helios, Its Next-Gen AI Powerhouse With MI455X & 6th Gen EPYC, Challenging NVIDIA’s Rack-Scale Dominance
1016 Comments -
NVIDIA GeForce RTX 50 SUPER GPUs Have Reportedly Arrived At AIBs, But Are On Hold Due To Undecided Memory Prices
787 Comments -
NVIDIA DLSS 5 Hands Over Full Control To Artists To “Direct The Final Frame”, As SIGGRAPH Technical Demo Shows How Neural Rendering Solved Big Challenge To Achieve 4K “Life-Like” Visuals On A Single GPU
622 Comments