NVDA Logs Best Day In Over A Month: Burry Spotlights Partnership Blitz — Analyst Calls Full-Stack Strategy ‘Underappreciated’

• Michael Burry spotlighted Nvidia’s partnership blitz, including Google Cloud’s plans to scale infrastructure to as many as 960,000 Rubin GPUs.
• Munster called Nvidia’s full-stack strategy underappreciated, citing higher switching costs and recurring upgrade cycles.
• Nvidia’s DSX Pod could shorten deployment from four to five months to about one month, reducing idle computing capacity.

Shares of Nvidia Corp. (NVDA) logged their best day in over a month on Wednesday as “The Big Short” investor Michael Burry spotlighted a wave of partnership announcements and analyst Gene Munster argued that investors still underappreciated the company’s full-stack AI strategy.

NVDA stock closed nearly 3% higher at $221.97 after a string of announcements at GTC Berlin, before slipping about 1.6% overnight early on Thursday.

Read Next

NOK Stock Heads For A Green Week After A 10% Jump In Q3 Sales As AI And Cloud Demand Drives Growth

Shivani Kumaresan

Michael Burry Flags Major Partnership Blitz

“Massive Partnership news day for NVDA,” Burry said in a subscriber chat on Cassandra Unchained. 

Google Cloud plans to launch general availability of its Vera Rubin-powered A5X instances in November, with European expansion planned. Its infrastructure could eventually scale to 960,000 Rubin GPUs across multiple sites.

Dell’s AI Factory with Nvidia now serves more than 6,500 customers, up from 5,000 last quarter. HPE expanded its enterprise AI offerings, while Nutanix and Netris announced infrastructure work for AI cloud providers. Meanwhile, CrowdStrike is integrating Falcon Guardian with Nvidia OpenShell and plans to add Sentry integration on BlueField-4 processors to enforce security policies in hardware.

Quantum X Labs reported improved error-correction results using Nvidia’s CUDA-Q software. Nvidia is also in talks about further investment in GPU cloud provider Firmus, The Information reported.

Beyond GPUs: Munster Sees A Lasting Advantage

Deepwater Asset Management managing partner Gene Munster said Nvidia’s integrated platform creates a lasting business advantage. “Software + libraries + hardware form a full stack that raises switching costs and supports recurring upgrade cycles,” he said on X, calling the strategy “Still underappreciated by investors.”

CEO Jensen Huang echoed the message: “NVIDIA today is really a full stack, end-to-end, complete AI factory platform.”

Huang said Nvidia had invested nearly $10 billion in Europe over the past year and established more than 12,000 partnerships across the region. Vera Rubin is in full production, he said, claiming 30 times Blackwell’s performance in large language model processing.

DSX Pod Promises Faster AI Deployment

Nvidia’s factory-assembled DSX Pod integrates computing, networking and cooling. Huang said it could cut the time from equipment arrival to token production from four to five months to about one month. “The wasted idle compute, the wasted idle CapEx will reduce, and everybody becomes more profitable,” he said.

Vertiv said its Nvidia DSX-based approach could produce up to 30% more cumulative tokens in the first 12 months from project kickoff under modeled assumptions, using the same installed capacity. Polar Data Centers is applying the approach to a 41-megawatt development phase in Norway.

AI Agents Send GPU Demand ‘Through The Roof’

“Agents think, reason, iterate, use tools, and as a result, consume an enormous amount of tokens,” Huang said. “GPU demand has really gone through the roof.”

Munster sees the trend sustaining computing demand. “Agents can consume 10-15x the tokens of a chatbot. Today roughly 15m consumers use bots daily. In a few years that could be billions,” he said.

Nvidia’s OpenShell controls agents’ access to files, networks and tools, with Sentry adding hardware-based monitoring. “I'm delighted to say that Anthropic and NVIDIA are partnering to create Claude OpenShell agents,” Huang said. He also expects software suppliers to become agent providers. Munster was more selective, saying that Salesforce and Adobe could struggle while security and data-monitoring companies benefit from increased activity and complexity.

Physical AI Offers Nvidia Its Next Growth Engine

Huang announced Hyperion-compatible vehicles from Volkswagen, JLR, Volvo and Mercedes. Munster estimated that physical AI could eventually contribute a third of Nvidia’s revenue, versus less than 5% today. Still, he cautioned that automakers face substantial development work. “Hyperion helps traditional automakers build autonomy, but it’s not a turnkey solution, they still lack the data and model,” he said.

Nvidia also introduced the Agile ONE reference robot with Agile Robots, expected in the first quarter of 2027, and RTX Spark, developed through a four-year Microsoft partnership. Huang said RTX Spark offers one petaflop of computing performance and 128 gigabytes of memory for local AI development.

The keynote included no new financial metrics or guidance, Munster noted. Instead, he highlighted SpaceX’s spending plans ahead of Nvidia’s next earnings report. “Last night’s $SPCX December quarter capex guide of $35-40B (street was at $19B) is a positive read-through for what we should hear from Nvidia on the earnings call in about a month,” he said.

How Do Retail Traders Feel About NVDA?

On Stocktwits, retail sentiment for NVDA slipped to ‘neutral’ from ‘bullish’ levels a day ago amid ‘normal’ message volume.

One user said, “$NVDA you guys remember when $230 was ”rock solid support”??  sorta shows ya how little you actually know.”

Another user said, “$NVDA DSX Pods are a real game changer. Plug and play.”

NVDA stock has risen 19% year-to-date.

For updates and corrections, email newsroom[at]stocktwits[dot]com.

Read Next: SPCX Stock Eyes Best Week In Nearly 3 Months — But Analyst Warns Investors May Have To ‘Wait Decades’ To Justify Its Valuation