Upstart chipmakers keep challenging Nvidia. This time it's Microsoft-backed D-Matrix

Cnbc · Jun 09, 2026 · 840 words · By Katie Tarasov

SRAM vs DRAM Technical Trade-offs Market Valuation and Investment AI Hardware Competition

headphones Listen to the eFinder podcast briefing

Ready to play

Daily briefing

What to know about SRAM vs DRAM Technical Trade-offs

In the increasingly competitive AI chip market, there's another startup in production that claims an advantage over Nvidia, the world's most valuable company.

Claims checked 22

Techniques found 1

Topics 3

Coverage spectrum

Coverage gap: Low Left coverage

Left0%

Center50%

Right50%

2 sources compared across this story cluster. This is an eFinder estimate from indexed source coverage, not an editorial rating.

What happened

In the increasingly competitive AI chip market, there's another startup in production that claims an advantage over Nvidia, the world's most valuable company.

Why it matters

D-Matrix, located three miles away from Nvidia's Silicon Valley headquarters, says its chips can run inference workloads 10 times faster and using five times less energy than a standalone graphics processing unit from the market leader — as long as the…

Common ground

The new inference chip, called Corsair, takes a novel approach to memory that's similar to Cerebras and Groq.

Perspective signals

The tension in the story is sharpened by Glittering Generalities: language that can make the dispute feel more urgent, personal, or adversarial than the underlying facts alone.

Follow-up questions

What new context would change how readers understand this SRAM vs DRAM Technical Trade-offs story?
What evidence would most clearly confirm or weaken the claim that D-Matrix begins shipping to those customers this month?
How does this story connect SRAM vs DRAM Technical Trade-offs with Market Valuation and Investment over the next few days?

open_in_new Read the original article: https://www.cnbc.com/2026/06/09/nvidia-d-matrix-chip-production-microsoft.html

psychologyPropaganda Techniques Detected

eFinder identified 1 propaganda technique in this article. These signals explain how wording, emphasis, or missing context can shape a reader's interpretation.

warning

Glittering Generalities 70% confidence

Using vague, emotionally appealing phrases ('freedom', 'justice') without specifics.

Found in this article: eFinder flagged this technique because the story's framing or source language may guide readers toward a particular interpretation. Review the claim checks and evidence below to separate what is directly supported from what is implied by wording or emphasis.

Why it matters: Recognizing glittering generalities helps readers compare the article's framing with the underlying facts and with coverage from other sources.

fact_checkClaims Checked

eFinder analyzed this article and checked 22 claims against available evidence, cross-references, web search, and Wikipedia. Here is what the fact-checking layer found.

schedule Pending 12

check_circle Corroborated 5

verified Verified By Reference 2

help Insufficient Evidence 1

verified Verified 1

info Single Source 1

help

Claim 1: “D-Matrix begins shipping to those customers this month.”

INSUFFICIENT EVIDENCE

No evidence was provided for this claim.

check_circle

Claim 2: “Cerebras, founded in 2015, held a blockbuster IPO last month, raising over $5.5 billion, and is now valued at over $50 billion.”

CORROBORATED

Multiple web search results confirm Cerebras had a massive IPO in 2026, raising approximately $5.5-5.6 billion with a valuation around $56 billion.

menu_book

wikipedia NEUTRAL — Andrew D. Feldman is an American technology entrepreneur best known as the co-founder and chief executive officer (CEO) of Cerebras Systems, a semiconductor and artificial intelligence infrastructure …
https://en.wikipedia.org/wiki/Andrew_Feldman_(businessman)

menu_book

wikipedia NEUTRAL — Arm Holdings plc is a British semiconductor and software design company headquartered in Cambridge, England, whose primary business is the design of central processing unit (CPU) cores that implement …
https://en.wikipedia.org/wiki/Arm_Holdings

menu_book

wikipedia NEUTRAL — Cerebras Systems Inc., headquartered in Sunnyvale, California, develops semiconductors, supercomputers, and related software to power artificial intelligence deep-learning applications such as inferen…
https://en.wikipedia.org/wiki/Cerebras_Systems

+ 3 more evidence sources

check_circle

Claim 3: “Groq's assets were bought by Nvidia for $20 billion in December”

CORROBORATED

Three independent cross-references from CNBC explicitly state that Nvidia acquired assets from Groq for $20 billion in December.

menu_book

wikipedia NEUTRAL — Steven Cliff Bartlett (born 26 August 1992) is a British entrepreneur, investor, and podcaster. In 2014, he founded Social Chain, and in 2017, started The Diary of a CEO podcast, which Spotify Wrapped…
https://en.wikipedia.org/wiki/Steven_Bartlett_(businessman)

menu_book

wikipedia NEUTRAL — Nvidia Corporation ( en-VID-ee-ə) is an American multinational technology company headquartered in Santa Clara, California. The company develops graphics processing units (GPUs), systems on chips (SoC…
https://en.wikipedia.org/wiki/Nvidia

menu_book

wikipedia NEUTRAL — Groq, Inc. is an American artificial intelligence (AI) company that builds an AI accelerator application-specific integrated circuit (ASIC). The architecture was originally introduced as a Tensor Stre…
https://en.wikipedia.org/wiki/Groq

+ 3 more evidence sources

verified

Claim 4: “Microsoft was one of the investors, through its M12 venture arm.”

VERIFIED

A web search result explicitly states that d-Matrix is an AI inference chip startup backed by M12, Microsoft's Venture Fund.

travel_explore

web search NEUTRAL — M12 is Microsoft's venture capital fund. We invest in early-stage technology companies disrupting the enterprise.
https://m12.vc/

travel_explore

web search NEUTRAL — Nov 15, 2023 ... “We were already looking for uses of generative AI before the whole thing exploded,” said Michael Stewart, a partner at M12, Microsoft's venture ...
https://news.crunchbase.com/ai/microsoft-m12-ai-investment-m…

travel_explore

web search NEUTRAL — Nov 12, 2025 ... d-Matrix, an AI inference chip startup backed by M12, Microsoft's Venture Fund, raised $275 million from new investors including Qatar ...
https://www.linkedin.com/posts/dina-bass_ai-chipmaker-d-matr…

schedule

Claim 5: “D-Matrix's Corsair chip achieves low latency inference on low power by tightly integrating memory and compute on a single chip.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 6: “Corsair the "densest SRAM solution in the market today," with up to 128 gigabytes of SRAM memory in a single server rack.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 7: “About 90% of them are in the U.S., while overseas customers are in the Middle East and Southeast Asia”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 8: “D-Matrix sells four Corsair chips packaged together inside a card that slides into slots in a data center server rack and costs tens of thousands of dollars”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

verified

Claim 9: “its chips can run inference workloads 10 times faster and using five times less energy than a standalone graphics processing unit from the market leader — as long as the workloads are small”

VERIFIED BY REFERENCE

The evidence provided for this claim is generic information about the letter 'D' and Nvidia's CUDA/GPU architecture. There is no specific evidence comparing D-Matrix chip performance (10x speed/5x energy) to Nvidia GPUs.

menu_book

wikipedia NEUTRAL — CUDA (Compute Unified Device Architecture) is a proprietary parallel computing platform and application programming interface (API) developed by Nvidia that allows software to use certain types of gra…
https://en.wikipedia.org/wiki/CUDA

menu_book

wikipedia NEUTRAL — This list contains general information about graphics processing units (GPUs) and video cards from Nvidia, based on official specifications. In addition some Nvidia motherboards come with integrated o…
https://en.wikipedia.org/wiki/List_of_Nvidia_graphics_proces…

menu_book

wikipedia NEUTRAL — PureVideo is Nvidia's hardware SIP core that performs video decoding. PureVideo is integrated into some of the Nvidia GPUs, and it supports hardware decoding of multiple video codec standards: MPEG-2,…
https://en.wikipedia.org/wiki/Nvidia_PureVideo

+ 3 more evidence sources

schedule

Claim 10: “GPUs rely on large amounts of another kind of memory called DRAM that's packaged into stacks of high bandwidth memory added around the logic chip.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 11: “When paired with an Nvidia Blackwell GPU, D-Matrix says, citing research from Gimlet Labs, that Corsair can run inference 10 times faster, three times cheaper and up to five times more energy efficiently than a standalone GPU.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

check_circle

Claim 12: “The new inference chip, called Corsair, takes a novel approach to memory that's similar to Cerebras and Groq.”

CORROBORATED

Multiple web search results confirm that D-Matrix (specifically the Corsair chip) uses a memory approach (SRAM/Digital Computing-in-Memory) similar to Groq and Cerebras.

travel_explore

web search NEUTRAL — Like Groq and Cerebras, D-Matrix relies on SRAM, a type of memory that can be made at logic fabs like Taiwan Semiconductor Manufacturing Company and integrated on the same chip.
https://www.entradaventures.com/post/upstart-chipmakers-keep…

travel_explore

web search NEUTRAL — Groq LPU uses only on-chip SRAM (no HBM), giving deterministic, jitter-free latency — the key differentiator for latency-critical applications. It is reportedly 10× more energy-efficient than GPUs for…
https://ai.plainenglish.io/ai-inference-architecture-the-mix…

travel_explore

web search NEUTRAL — Within CIM: d-Matrix Corsair uses Digital Computing-in-Memory (DIMC) as its core technology, embedding the AI model’s underlying architecture into the chip, applicable to all AI models with even great…
https://insights.trendforce.com/p/ai-inference-chip-architec…

schedule

Claim 13: “D-Matrix also teamed up with Arista, Broadcom and Super Micro to build a full rack-scale system called SquadRack”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

info

Claim 14: “Founded in 2019, D-Matrix has raised around $500 million so far, putting it at around a $2 billion valuation.”

SINGLE SOURCE

The evidence provided for this claim is completely irrelevant, discussing the movie 'The Matrix', Genesis AI, and Lionel Messi. No data on D-Matrix's founding date, funding, or valuation was provided in the evidence.

travel_explore

web search NEUTRAL — The movie depicts a dystopian future in which reality as perceived by most humans is actually a simulated reality called "the Matrix", created by sentient ma...
https://www.youtube.com/watch?v=6T_Rj47nm0Q

travel_explore

web search NEUTRAL — Genesis AI is negotiating to raise approximately $500 million, potentially valuing the startup at $3 billion. Recently emerged from stealth, the company specializes in universal robotics models.
https://aihunt.app/shorts/genesis-ai-in-talks-for-500m-fundi…

travel_explore

web search NEUTRAL — Spanning approximately 10,500 square feet, the contemporary property includes 10 bedrooms, nine bathrooms, expansive living spaces and floor-to-ceiling glazing that opens onto views of the water.
https://www.houseandgarden.co.uk/article/where-does-lionel-m…

schedule

Claim 15: “D-Matrix's next chip, Raptor, is scheduled to launch next year on TSMC 4 nanometer”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 16: “That DRAM is also what's in short supply from Micron, Samsung and SK Hynix.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 17: “D-Matrix relies on SRAM, a type of memory that can be made at logic fabs like Taiwan Semiconductor Manufacturing Company and integrated on the same chip.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

schedule

Claim 18: “Nvidia CEO Jensen Huang said last week that his company remains the leader in low-cost inference with its leading Vera Rubin system”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

check_circle

Claim 19: “Nvidia then released a new Groq chip at GTC in March, called a language processing unit.”

CORROBORATED

Multiple sources confirm that Nvidia released a language processing unit (LPU) based on Groq technology at GTC in March.

travel_explore

web search NEUTRAL — Groq created a chip it called a language processing unit (LPU), used for inference, and sold it as part of a cloud service or an on-premises hardware cluster. With Nvidia now owning the IP for LPUs, t…
https://techcrunch.com/2026/06/22/ai-chipmaker-groq-confirms…

travel_explore

web search NEUTRAL — NVIDIA confirmed a $20 billion licensing deal with Groq to build a dedicated AI inference chip using LPU technology. OpenAI is the primary customer. The new chip debuts at GTC on March 16.
https://udit.co/blog/nvidia-20-billion-groq-deal-inference-c…

travel_explore

web search NEUTRAL — Nvidia then released a new Groq chip at GTC in March, called a language processing unit.
https://www.bundle.app/en/technology/upstart-chipmakers-keep…

schedule

Claim 20: “The chip is made in Taiwan on TSMC's 6-nanometer node.”

PENDING

This claim was extracted as a checkable statement from the article. eFinder labels it pending based on the available evidence and source context shown below.

check_circle

Claim 21: “Microsoft's own chip ambitions, including its Maia 200 chip for AI inference, new PC processors built with Nvidia, and an in-house quantum computing chip announced last week.”

CORROBORATED

Multiple sources confirm the existence and unveiling of the Microsoft Maia 200 AI inference chip.

travel_explore

web search NEUTRAL — Microsoft discussed its Maya 200 AI inference chip with data on performance benchmarks. 00:08:29. . The chip delivers three times the FP4 performance of Amazon's third-generation Tranium chips and exc…
https://www.podfolio.co/podcast/machine-learning-street-talk…

travel_explore

web search NEUTRAL — Microsoft Unveils Maia 200 AI Inference Chip Built on 3nm Process Microsoft has unveiled the Maia 200, a cutting-edge 3nm AI inference chip designed to deliver unmatched performance per dollar and wat…
https://www.linkedin.com/posts/pooja3p_microsoft-azure-aiinf…

travel_explore

web search NEUTRAL — The Great AI Chip War: Microsoft Challenges NVIDIA Dominance. Microsoft’s announcement signals the latest escalation in the intensifying battle for AI hardware supremacy. For years, NVIDIA’s GPUs have…
https://cryptorank.io/news/feed/200d1-microsoft-maia-200-ai-…

verified

Claim 22: “D-Matrix, located three miles away from Nvidia's Silicon Valley headquarters”

VERIFIED BY REFERENCE

The provided evidence for this claim consists of generic Wikipedia entries for the letter 'D' and unrelated topics. There is no information regarding D-Matrix's physical location relative to Nvidia.

menu_book

wikipedia NEUTRAL — The Silicon Genesis Project is an on-going oral history project on the history of Silicon Valley and the semiconductor industry, conducted by volunteers and housed at the Stanford University Library.
https://en.wikipedia.org/wiki/Silicon_Genesis_Project

menu_book

wikipedia NEUTRAL — Tensor Processing Unit (TPU) is a neural processing unit (NPU) application-specific integrated circuit (ASIC) developed by Google for neural network machine learning. Tensorflow, Jax, and PyTorch are …
https://en.wikipedia.org/wiki/Tensor_Processing_Unit

+ 3 more evidence sources

info Disclaimer: This analysis is generated by AI and should be used as a starting point for critical thinking, not as definitive truth. Claims are verified against publicly available sources. Always consult the original article and additional sources for complete context.