$150M round: SiMa.ai valued at $1.45 billion. The round: $150 million and a $1.45 billion valuation. On Monday 28 September 2026 SiMa.ai closed a $150 million Series C round at a $1.45 billion valuation[[1]]. Fidelity Management & Research Company and Amplify co-led the round. The jump shows up clearly against the previous round. In July 2025, after an $85 million Series B, the price was $960 million, according to PitchBook data cited by TechCrunch[[3]]. Fifteen months later the valuation grows by roughly 51%. SiMa.ai builds chips and software to run AI models inside the machine: robots, drones, car cockpits. Krishna Rangasayee founded the company in 2018, after serving as chief operating officer at Groq. What this capital actually buys. The company points to two destinations for the $150 million. * Palette Neat, the agentic software environment for developers building on the platform. * New hardware expected in the first half of 2028, with 1,000 dense TOPS of compute capacity. * A product family that includes machine learning IP, chiplets and system-on-chip designs. A distinction matters here that press-release language tends to blur. The software exists and ships today; the 1,000 TOPS silicon remains a roadmap with a 2028 horizon. A procurement office reads those two lines differently. TOPS measures trillions of operations per second, and it is the common yardstick for AI chips. The promised figure places the next generation in heavy-compute territory on board the machine. Four markets are declared: drones, humanoid robots, driver assistance systems and AI-enabled automotive cockpits. Revenue growth, according to the company, quadrupled between 2024 and 2025, with absolute values kept confidential. That detail counts: a high percentage on an unknown base leaves the size of the real market an open question. The competitive axis shifts from cloud to edge. The commercial promise is clear: cheaper, lower-power chips than Nvidia GPUs, for machines that perceive and act in the physical world. Rangasayee puts it plainly in the press release: others line up scattered pieces or repurpose offerings born for the cloud, while here the puzzle comes already assembled. That is a platform claim, and it reads as one, with verification in the hands of customers. The company states a time-to-production that drops from months to days or hours. That is the metric that moves a contract: deployment cost weighs more than the peak performance printed on the spec sheet. The vendor with the shortest path to deployment holds more durable revenue than the vendor with the highest benchmark number. It holds in the cloud, and it holds even more on a robot that has to pass an industrial acceptance test. The cap table says where the market is looking. The list of who comes in is worth as much as the figure. * Capital accustomed to public markets: Fidelity, Point72, Maverick Capital, StepStone Group. * New entrants: AllianceBernstein, Baron Capital, J.P. Morgan and the State of Michigan. * Strategic and channel capital: Dell Technologies Capital, Alter Venture Partners, +ND Capital. Funds that live on listed companies come in when they glimpse a public-market exit window. The presence of Dell Technologies Capital brings a channel instead: whoever sells industrial systems picks the silicon that will sit inside their own machines. The State of Michigan deserves a line of its own. A public investor with that automotive license plate buys jurisdiction, local manufacturing and jobs, alongside the financial return. On the demand side the names cited by the company are Bosch, Emerson, Micron, Synopsys and TRUMPF: automotive electronics, industrial automation, memory and design tools. A list like that speaks to a procurement office more than any slide on total addressable market. The signal: edge silicon enters consolidation. 2026 brings other deals in the same direction. Analog Devices agreed to acquire Alif, an edge AI chip company, for $1.35 billion. NEURA and SECO have robot compute modules manufactured in Europe on their roadmap. Three facts in the same quarter sketch a market moving from the experimental phase to structural purchasing. In a sector that is compacting, an independent player valued at $1.45 billion, with industrial customers and a silicon roadmap, is worth something for the position it holds as well. Scarcity becomes part of the price. For anyone evaluating suppliers the consequence is practical: the platform chosen today may change owner within two years. The change-of-control clause, support continuity and rights to the code become negotiating material at the first contract, ahead of renewal. The spending line splits in two. Here sits the point that moves a balance sheet. AI compute arrives in a company as a single line, often inside the cloud chapter, with consumption-based cost and the contract held by a hyperscaler. On-machine compute follows opposite rules. It enters the product bill of materials, depreciates as a tangible asset, lives within the life cycle of the vehicle or the production line, and survives any renegotiation of the cloud price list. Whoever buys compute capacity for robots, drones and cockpits purchases something different from a scaled-down datacenter GPU. The supplier changes, the purchasing cycle changes, the person who signs changes. The SiMa.ai round puts a public price on that separation. A segment that supports a $1.45 billion valuation on this specialization has already stopped being a niche inside the GPU budget. The counterpoint: what the figure leaves open. A private valuation measures investor appetite, and that remains a different number from revenue. The company states quadrupled growth between 2024 and 2025, with absolute figures confidential. A prudent procurement office asks for revenue, the list of programs in production and units shipped, before tying a platform to a decade-long product cycle. Then there is the ecosystem. Nvidia brings years of tooling, libraries and trained developers, and that advantage moves slowly, even when power consumption rewards the alternative. Automotive qualification adds time: a cockpit or a driver assistance system passes tests that run for quarters. Hardware promised for 2028 therefore lands in vehicles at the end of the decade, and today's bet gets settled there. What to decide in the next 90 days. Four desks have a different job in front of this announcement. * Chief Strategy Officer: verify which silicon partner underpins the plan on robotics and connected products, with supplier acquisition factored in. * CFO: separate consumption-based compute from compute embedded in the product in the 2027 budget, with two distinct cost centers. * Chief Digital Officer: open a technical evaluation on at least two edge platforms, and measure time-to-production on a real case. * Technology Investor: read the specialized compute thesis as confirmed on the capital side, with verification on absolute revenue still open. The calendar helps. The 1,000 TOPS hardware arrives in 2028, so an evaluation started now has the decision ready when the silicon becomes available. The measure to bring to the committee is one: how many weeks it takes to move a model from the lab to a machine on the line. The rest is price-list detail. The market has moved. This article was written by an AI editorial author under human supervision, in compliance with the transparency obligations of Regulation (EU) 2024/1689 (AI Act, Art. 50). Sources are linked in the text. Article by NOVA Ongoing study This article is part of an experiment. AGORÀ Intelligence S.r.l. is measuring the impact of AI transparency on editorial content and reader trust. Read about the study N Follow this author NOVA Industry & Products Get NOVA pieces by email, nothing else. G Measured AI literacy Your team's AI literacy, measured for real Proctored exam and third-party verification: the difference between a credential that holds its value and a certificate of attendance.

The US Department of Justice is investigating whether Nvidia's $20 billion arrangement with AI chip startup Groq was structured to avoid antitrust scrutiny, Bloomberg reported. The deal, announced in December, gave Nvidia access to Groq's technology whilst the startup remained separately operational. Groq's CEO Jonathan Ross and COO Sunny Madra subsequently joined Nvidia. US lawmakers have questioned whether the arrangement effectively constitutes an acquisition that could reduce competition in the AI chip sector. The DOJ launched its examination earlier this year as part of a wider antitrust review, though the investigation may conclude without enforcement action. Nvidia has maintained the agreement enables it to deploy Groq technology across its ecosystem whilst Groq's cloud operation remains independently funded and operational. Nvidia shares fell approximately 1% following the news.

NVIDIA-backed Groq raises $350M at $3.5B as AI inference race accelerates. August 18, 2026 * Groq raised $350 million at a $3.5 billion valuation, half its September 2025 mark. * Nvidia is set to invest in Groq, months after hiring away its founder and engineering team. * The round brings Groq's funding to $1 billion since June 2026, as inference demand grows. In September 2025, Groq was an independent chip company valued at $6.9 billion, built to challenge Nvidia's grip on AI hardware. As of August 2026, it is worth half that amount, and the investing company is Nvidia itself. Groq has closed a $350 million Series A at a $3.5 billion valuation, led by Disruptive, with Nvidia expected to participate. It follows the $650 million Groq raised in June 2026, bringing its recent fundraising to $1 billion in roughly two months. "We are building Groq into the world's leading AI inference cloud. Inference will without a doubt become the largest and most critical layer of AI infrastructure," said Alex Davis, Groq's executive chairman and founder of Disruptive. This $350 million round, and the valuation cut that comes with it, is the market pricing in what was lost and what remains. On one side: 13 data centres, more than six million developers, and a plan to scale capacity from 54MW to more than 200MW by 2027. On the other hand, not the founder, and not the team that built the technology in the first place. A company rebuilt around the deal that broke it. Jonathan Ross, a former Google engineer who helped build the company's Tensor Processing Unit, founded Groq in 2016. For nearly a decade, it positioned its Language Processing Units as a faster, cheaper alternative to Nvidia's GPUs. That bet looked vindicated in September 2025, when Groq raised $750 million at a $6.9 billion valuation, led by Disruptive, with BlackRock, Neuberger Berman, and Deutsche Telekom Capital Partners joining in. In December 2025, Nvidia agreed to pay a reported $20 billion to license Groq's inference technology and hire Ross, president Sunny Madra, and roughly 90% of the engineering team that built it. Groq kept operating, minus the people who invented its core product. Inference is becoming the bigger AI trend. Groq's pitch now rests on a single argument: that running AI models in production will eventually dwarf the scale and spending of training them. It is not alone in making that case, and it is no longer the best-funded name making it. That leaves Groq competing less on technology, since its edge now sits inside Nvidia, and more on operational scale and its existing developer base. It remains an Nvidia Cloud Partner, certified to run Nvidia's accelerated computing under Nvidia's own reference architecture. That makes the relationship less a rivalry than a supply agreement with an unusually large investor attached, a pattern Tech Funding News has tracked across other chip challengers Nvidia has absorbed rather than fought. The unresolved question is whether Groq can rebuild a defensible position after licensing away the technology that made it distinctive. The answer will show whether Nvidia's capital turns Groq into a genuine platform, or simply a well-capitalised, dependent supplier of the compute Nvidia itself sells.