SUCCESS

The Democrats who may get a bigger say in future crypto legislative efforts are familiar figures, and they generally look at digital assets with distrust.

Total deposits grew from $1.1 billion in March to $4.6 billion by July, driven by clients in crypto, AI and defense.

‘Reacher’ is back this week with season 4, and its Rotten Tomatoes review score from critics is certainly eyebrow-raising.

Many of today’s enterprises have a shoebox problem: fragmented, incomplete, and duplicated records, forcing customers or employees to connect the dots themselves.

Mistral AI wants to turn European AI sovereignty from a talking point into a product — one with a service-level agreement attached.The French artificial intelligence company announced Tuesday a three-part expansion of its infrastructure business: regional inference endpoints that let customers choose whether their AI workloads run in Europe or the United States, a new “Priority Tier” backed by an uptime guarantee for mission-critical deployments, and a coalition of European enterprises making multi-year compute commitments that Mistral says will underwrite 200 megawatts of infrastructure across Europe by the end of 2027 — and a full gigawatt by the end of 2030.In a move that may raise eyebrows among sovereignty purists, the company also said it will begin hosting third-party open models on its platform, starting with GLM-5.2 from Z.ai, the Chinese AI lab formerly known as Zhipu.Taken together, the announcements mark a decisive shift in how Mistral positions itself. The company that built its reputation training open-weight language models is now selling something closer to critical infrastructure: assured capacity, regional control, and contractual reliability for enterprises and governments that want frontier AI without surrendering control over where it runs.”When we spoke in June, the story was around how Mistral was building a full-stack AI offering,” Timothée Lacroix, Mistral’s co-founder and chief technology officer, told VentureBeat in an exclusive interview ahead of the announcement. “Today, the announcement is about strengthening one part of this infrastructure, which is the inference part.”That one part, it turns out, comes with a price tag measured in the tens of billions of dollars.Inside Mistral’s plan to build 1 gigawatt of European AI compute by 2030The headline numbers deserve scrutiny, because they imply staggering capital requirements. Mistral currently operates less than 200 megawatts of capacity, according to the company. Details shared with VentureBeat show the near-term buildout resting on three sites: a 44-megawatt facility near Paris that became operational in the second quarter of this year, a 23-megawatt facility in Sweden built in partnership with EcoDataCenter using renewable energy and advanced cooling, and a 10-megawatt site in Les Ulis, France, that came online in the third quarter.Getting from there to one gigawatt by 2030 is a different order of magnitude. Independent estimates suggest just how different: research firm Epoch AI calculates that a typical one-gigawatt AI data center requires roughly $38 billion in upfront capital expenditure, with servers and GPUs — not buildings or land — consuming the majority of the cost. Goldman Sachs Research pegs next-generation AI facilities at $15 million to $20 million per megawatt before accounting for the chips inside them.Lacroix did not dispute the scale of the challenge. The investment required for a gigawatt of capacity “is a large investment that requires also a lot of scaling and revenue behind it,” he said.The urgency, in his telling, comes from a supply crunch that is about to get worse. “More and more, and especially around 2027 and 2028, we see that the demand for AI compute is exceeding what the market has to offer, especially in Europe,” Lacroix said. McKinsey has estimated that meeting global AI demand could require $5.2 trillion in data-center capital expenditure by 2030 — and Europe, by most analyses, is starting from behind.A company valued at a fraction of its American rivals cannot close that gap with venture capital alone. Which explains the most consequential — and most unusual — piece of Tuesday’s announcement.European Compute Units turn AI sovereignty into a five-year contractMistral is assembling what it calls an anchor group of enterprises whose long-term commitments will collectively finance infrastructure none of them could justify alone. Those commitments convert into “European Compute Units,” or ECUs — a claim on Mistral-built capacity over multiple years that participants can spend on inference, training, model adaptation, or other AI workloads as their needs evolve.If that structure sounds more like a power-purchase agreement than a cloud contract, that appears to be the point. Data-center financing increasingly resembles large infrastructure projects — gigawatts, substations, energy agreements — rather than traditional technology spending, and lenders want demand locked in before capital gets deployed. Mistral raised €830 million ($962 million) in debt earlier this year to fund its data center near Paris, TechCrunch reported in March, and pre-committed enterprise demand is exactly what makes that kind of financing repeatable at ten times the scale.Lacroix was unusually direct about the mechanics. “The entire point of compute units is to have commitment,” he said. “The goal is to have customers commit for around five years, or at least a long time.” Asked what happens if a customer wants out early, he didn’t soften the answer: “There is no getting out.”What makes a five-year, no-exit commitment palatable, he argued, is flexibility in how the capacity gets consumed. “Typically this can be spent on raw inference that you then feed through any other AI stack. It can be spent on raw compute as managed Kubernetes, and it can be spent at the very top with our full AI offering,” he said. “My hope is that they will use it with our full-stack services and will love it.”The anchor group already includes some of Europe’s industrial heavyweights. Amadeus CEO Luis Maroto said in a statement that “capacity, deployment control, and operating continuity become increasingly important for all enterprises.” ASML chief Christophe Fouquet — whose company led Mistral’s $13.4 billion (€11.7 billion) Series C last year — called building European AI capacity one of the few industrial endeavors that “will matter more to Europe’s next generation,” while Capgemini’s Aiman Ezzat framed it as “a question of who shapes the future of European industry.” CMA CGM chairman Rodolphe Saadé said the shipping group’s Mistral deployment is “already under way among thousands of employees.”Commitments of that duration only make sense, of course, if the sovereignty being purchased is real. On that question, Mistral’s announcement contains an asterisk worth reading closely.The fine print on sovereign AI: what data can still leave EuropeThe centerpiece product is Mistral Regional Endpoints, now generally available, which let customers pin inference and its associated processing to Europe or the U.S. Alongside it, the new Priority Tier — in public preview — offers committed service levels, custom rate limits, and an uptime SLA for mission-critical workloads.Mistral claims it is the only European AI lab offering both a choice of processing region and an SLA-backed service tier, and Lacroix said a third option is coming: an endpoint “that stays on Mistral-controlled infrastructure, so on Mistral compute” — for customers who want their inference not just in Europe, but off hyperscaler hardware entirely.Then comes the fine print. Mistral’s own materials note that in-region inference remains subject to “limited, safeguarded transfers” to sub-processors that may sit outside the chosen region. Pressed on what actually leaves Europe, Lacroix pointed to the connective tissue of modern AI applications: tool calls.”There are some tool services, like some tool calls, that might be hosted in places where we don’t fully control this,” he said, citing web search as an example. “A few of our web-search providers might not all be in Europe, and in that case, we need to potentially gate that capability.”His answer to the compliance question — would this satisfy a European bank or a defense ministry? — was that gating is the feature, not the bug. Capabilities that cannot be sourced in-region can be switched off entirely, restricted to certain users or workspaces, or, given sufficient demand, rebuilt with European providers. “Any capabilities that we don’t find a provider for in Europe — if it needs to be done in Europe, we’ll find some way to implement it or find ways to address it,” Lacroix said.For enterprise buyers, that is a more honest framing than most sovereignty marketing offers: full regional control is available, but the moment an AI agent reaches out to the open web, sovereignty becomes a configuration decision rather than a default. The same pragmatism runs through the announcement’s most surprising line item.Why Europe’s open source AI champion is hosting China’s GLM-5.2A French national champion — one that has partnered with the French army and positioned itself as Europe’s answer to American AI dependence — hosting a Chinese lab’s model invites an obvious question. Lacroix’s answer was disarmingly matter-of-fact.”It’s a great model. Everyone loves it. It’s open weight, so there was no good reason for us not to do it, really,” he said, noting that Mistral’s own stack is already built on open-source software like Kubernetes.On security vetting, he argued that open weights fundamentally change the risk calculus. “The risks in taking a new model, at the layer of the weights, are — at least in my opinion — rather limited,” Lacroix said. “We checked basically all of the safety and compliance evals that we have. We’ll control that model, its outputs, and what it does the same way we do any of our models. We have the same inputs and outputs and monitoring capabilities over all of it.”The strategic logic is worth unpacking. By hosting third-party open models under European regional controls and the same SLAs as its own, Mistral is repositioning itself from model vendor to sovereign distribution layer — the trusted intermediary through which any open model, regardless of origin, can be consumed by a regulated European enterprise that could never call a Chinese API directly. It is the “model garden” playbook the hyperscalers run with Bedrock and Vertex, executed on European soil with European guarantees.Customers appear to be reading it that way. “Mistral allows us to run open models under strict regional controls and service commitments, making it easy for us to maintain data residency and compliance requirements,” Matan Griberg, CEO of AI software-engineering company Factory, said in a statement.Lacroix stressed the move is not a retreat from frontier training: the model Mistral had in training as of June “is still training, and we’re still very excited about it,” he said. But openness to rivals’ models signals where the company now believes its moat lies — not in any single model, but in the infrastructure underneath all of them. Which makes its relationship with the world’s most powerful infrastructure company all the more interesting.How the multibillion-dollar Microsoft deal funds Mistral’s independenceHovering over every sovereignty claim is Mistral’s deepening relationship with Microsoft. In July, the two companies announced a multibillion-dollar expansion of their partnership under which Microsoft will rent capacity from Mistral’s European data centers to serve its own cloud and AI demand, while adding Mistral Medium 3.5 and OCR 4 to Microsoft Foundry, bringing Medium 3.5 to Copilot Studio, and enabling Mistral models on Azure Local for disconnected, customer-controlled environments. Mistral CEO Arthur Mensch told The Wall Street Journal at the time that two-thirds of Mistral’s customers already work with Microsoft.How does a company selling independence from U.S. hyperscalers square taking one on as its largest tenant? Lacroix described Microsoft not as a patron but as an anchor customer that de-risks the buildout.”It allows us to scale different parts of the business differently by building infrastructure with Microsoft as a customer,” he said. “We can scale that team, we can scale our infrastructure, and make sure that we can then, on the side of it, also build for ourselves and for our customers.” He compared the arrangement to the neocloud playbook — companies that built businesses supplying capacity to the hyperscalers themselves. “As that part of our business resembles that of neoclouds, we’re following the same thing.”It is a genuinely clever inversion: rather than renting American infrastructure, Mistral is renting infrastructure to one of America’s largest companies, using Microsoft’s demand to finance capacity that also serves European sovereignty customers. But the independence has limits no contract can engineer away — the GPUs filling Mistral’s European data centers come overwhelmingly from Nvidia and other American chipmakers, as SiliconANGLE noted in its coverage of the July deal.Asked directly why a customer should choose Mistral over an EU region on AWS or Azure, Lacroix gave two answers. “The simplest possible answer is capacity. There is more demand than supply right now, and so it adds another option,” he said. The second cuts closer to the pitch: “We are a European provider, and on the region that would be Mistral compute, we are fully independent. That’s a truly differentiated offering than all of the hyperscalers or pure inference companies can provide.”The economics of open models: why agentic AI is pushing inference to the cloudThere has always been a tension at the heart of Mistral’s business: its best-known models are free to download, and open models have historically been difficult to monetize through APIs. Asked how free weights fund a gigawatt buildout, Lacroix offered the clearest articulation yet of the company’s thesis — that the economics of self-hosting are collapsing under the weight of the models themselves.”When the models were smaller, and we were before the explosion of agentic AI, it was doable for enterprises to host their own — up to, let’s say, 100-billion-parameter dense models — on their premises,” he said. “More and more, with models going into the trillion or more parameters, with the current hardware, and with the increasing amount of tokens that need to be processed, it becomes harder.”His conclusion was blunt: “I don’t see how, with the current trend of model size and growth of agentic tokens, we keep the full inference on-prem. To me, that is why we think we’re going to monetize our cloud inference.” Inference, he noted, is particularly well suited to the cloud because it “does not need to hold any data” and can be encrypted in transit.In other words: open weights get Mistral into the enterprise, and the physics of trillion-parameter agentic workloads brings the inference — and the revenue — back to Mistral’s data centers. The thesis will get an expensive test. Mistral has raised roughly $4 billion to date, according to PitchBook data — a fraction of the war chests assembled by OpenAI and Anthropic — and Bloomberg reported in June that the company is in talks to raise about €3 billion at a roughly €20 billion valuation, nearly double its Series C mark. The revenue behind the buildout will have to come from exactly the enterprises Tuesday’s announcement is courting.And Europe, in Mistral’s telling, is only the first market for what it is selling. Asked whether the framework could be replicated in the Middle East, Asia, or anywhere else anxious about AI dependence, Lacroix didn’t hedge: “It’s completely right. We’re starting this in Europe because it’s also an easier part of the world for us to scale into, especially in the infrastructure. But we definitely want to extend this, depending on customer demand.” Every layer of the stack, he said, “can be controlled, changed, replaced depending on where we operate and what the requirements are — that’s pretty much where we excel.”That is the wager underneath the SLAs, the compute units, and the Chinese model flying a European flag: in a world where the U.S. and China dominate frontier AI, the durable business is selling everyone else control. To fund it, Mistral is asking Europe’s largest enterprises to sign five-year contracts with no exit — while making a bigger, longer commitment of its own. A gigawatt, after all, is a promise measured in decades. For Mistral, too, there is no getting out.

The company doesn’t currently have a fix for the problem.

Suze Orman has warned for years that retreating from equities too early creates a risk most retirees fail to anticipate until the damage is done. Her argument targets an assumption baked into the old retirement playbook, one that newer industry research and professional practice now sharply contradict.A CNBC report published on August 8 revealed that mainstream financial advisors now recommend retirees hold 40% to 80% of their portfolios in stocks. That range marks a sharp departure from the older rule of thumb, which capped stock exposure at about 30% for newly retired investors.Orman has also targeted a costly product she believes compounds the problem, especially for people approaching retirement or already living on their drawdown. The question the industry shift raises is whether current retiree allocations can sustain purchasing power over 25 to 30 years of withdrawals.Financial advisors have rewritten the retirement allocation playbookFor decades, a standard guideline told retirees to reduce their stock allocation to 30% or less the moment they began drawing on their savings. The advisory profession has shifted well beyond that framework, Cheri Belski, head of investment management solutions at LPL Financial in Fort Mill, told CNBC.Belski told CNBC that the older 30% ceiling was “a general rule of thumb to reduce the equity portion of a portfolio as soon as you retire,” a benchmark advisors now consider outdated.That framework fit an era of higher bond yields and shorter retirements, when 85% of private-sector workers with plans held pensions in 1975, according to the National Institute on Retirement Security (NIRS).A 65-year-old retiring today could live another 30 years, the Social Security Administration (SSA) data showed, and bond-heavy portfolios often struggle to preserve purchasing power over that span.Orman’s long-standing case against going too conservative too earlyOrman has argued against excessive retirement conservatism for the better part of a decade, well before the mainstream advisory community caught up to her position.Orman restated her case for equity exposure on the January 11, 2026, episode of her Women & Money podcast, arguing that retirement allocations should be driven by financial situation rather than age. Her framing echoes the line she gave Money magazine in a March 2020 interview: “You’re retired. Your portfolio isn’t.”More Retirement:Retirement Tech in 2026: AI, Operational Efficiency, and Better Participant ExperienceGeorge Kamel, Rachel Cruze warn about a mortgage retirement trapMassachusetts retirement taxes explained: What retirees should know before moving or stayingOn her Women & Money podcast, Orman has directed her sharpest criticism at variable annuities purchased inside individual retirement accounts and 401(k) plans.She has called that combination one of the most expensive safety plays a retiree can make with their long-term savings.“It makes no sense for you to put a tax deferred investment such as an annuity within a tax deferred or tax free,” Orman said on the podcast.

Orman has long warned against overly conservative retirement portfolios, arguing that retirees still need equity exposure and growth potential.Joe Kohen / Getty Images

Variable annuity fees and commissions work against retirement savingsHer concern with these contracts inside retirement accounts centers on layered fees that compound over time and steadily reduce the capital available for growth. On the podcast, Orman estimated that the mortality charge for the death-benefit guarantee typically runs between 1.2% and 1.5% a year alone.That figure lands before investors account for underlying fund expenses, administrative charges, and optional riders that push total annual costs above 2% in many contracts.What variable annuity costs often includeMortality and expense risk charges, typically 1.2% to 1.5% annually, layered on top of underlying fund management fees, the Orman podcast showed.Surrender penalties for early withdrawals that often apply for six to eight years or longer after the purchase date, the U.S. Securities and Exchange Commission stated.Upfront advisor commissions that commonly range from 4% to 7% of the amount moved into the contract, Orman noted on her show.Stuart Katz, chief investment officer of Robertson Stephens in San Francisco, told CNBC that retirees need equities to address longevity and inflation risk. You need a portfolio allocation that has long-term growth benefits, and equities can serve that purpose, to address longevity risk and inflation.The SEC reinforces that point in its investor bulletin on variable annuities, stating that a variable annuity purchased inside a 401(k) or IRA offers no additional tax advantage. Fee-only planner Allan Roth made a similar point in a Forbes column, citing high fees and commission structures that can create conflicts of interest.The pressures changing the retirement allocation mathInflation and time working together against a portfolio that has stopped growing drive the shift toward higher equity allocations. BlackRock’s 2026 Income Outlook warned that money-market yields are falling as central banks cut rates, eroding the income retirees can generate from cash.The average American now believes they need $1.46 million to retire comfortably, up more than 15% from the prior year’s estimate, Northwestern Mutual reported. Nearly 48% of respondents said it is somewhat or very likely they will outlive their savings. Orman does not reject safety entirely, and she has recommended on her podcast that retirees hold three to five years of living expenses in cash, Yahoo Finance reported.That buffer sits apart from the invested portfolio, giving stocks and bonds time to recover from temporary declines without forced selling during market downturns.Where Orman’s argument and the revised advisory consensus convergeOrman’s core argument and the revised advisory guidance converge on one idea: the real danger is not short-term volatility but slow loss of purchasing power. The advisory shift reflects a longer expected retirement horizon and the compounding impact of inflation on cash-heavy portfolios, factors the older 30% guideline did not weigh.Belski and Katz focus on how much of a portfolio belongs in stocks, while Orman focuses on how product fees eat into long-term returns.All three land on the same underlying point: allocation decisions in a retirement account have more consequence over a 30-year drawdown than under the older, shorter-retirement assumption.Related: Americans face a painful hit to retirement in their 30s

Time and again, I’ve seen the same movie at Intel, from different angles. From the layoffs, the 18A yield progress, the management restructuring, the CHIPS Act subsidies, and the Q2 beat, which was the company’s strongest revenue growth in 15 years.Each piece of the puzzle has been pointing in the same direction: Artificial Intelligence (AI). These are just a few of the moves Intel has made. Recently, we saw Intel raise its 2026 capital expenditure estimates to more than $20 billion, from around $18 billion, to meet rising demand.August 10, Intel revealed how it intends to fund the next chapter. Intel announced plans to offer $15 billion in new stock. This may be its first public equity offering since it listed in 1971, according to Intel’s statement. The proceeds are earmarked for general corporate purposes, including real-world AI applications, purpose-built silicon, and foundry expansion.INTC ranks 5th among all S&P 500 components year-to-date, up 170.79%, according to Slickcharts data. It trails only SanDisk, Dell, Micron, and Seagate. Intel is no longer the turnaround story the market was skeptical about. It is actually one of the best-performing large-cap stocks in America, and management is already raising $15 billion to stay ahead of the AI demand wave driving that performance.Also Read: Intel Corporation Latest News and StoriesHere is why exactly Intel needs $15 billionThe capital raise addresses two specific needs simultaneously, and understanding both explains why management chose equity over debt.First, the foundry business. Intel has begun high-volume manufacturing of select Intel Core Ultra Series 3 “Panther Lake” processors using ASML’s high-NA extreme ultraviolet (EUV) lithography technology, making it the first company to ship high-volume logic products manufactured with the advanced technology.A €5 billion investment was announced to expand Xeon 6 manufacturing capacity. Intel expanded the Bowers campus to support current and future leading-edge process technology, according to Q2 fiscal 2026 financial results.More Intel:5-star analyst aggressively resets Intel stock price target after earningsGoldman delivers a candid response after Intel’s stunning quarterIntel makes another painful move in one of its key businessesJust for your information, building and equipping semiconductor cleanrooms is, by a large degree, one of the most capital-intensive activities in any industry. The $15 billion keeps that buildout funded without adding leverage to a balance sheet that has already been meaningfully repaired.Second, the AI inference and physical AI opportunity. Intel’s Q2 Data Center and AI segment revenue grew 59% year-over-year to $6.3 billion.Related: Intel’s 14A chips aren’t built yet: Synopsys is already thereIntel’s Q2 report states that more than 130 customers are currently adopting or testing Intel Core Ultra Series 3 and Core Series 3 processors for edge AI and robotics applications. The Vector Core Compute initiative, which combines Intel Xeon, SambaNova RDUs, and Nvidia Blackwell GPUs, is live. The inference opportunity is real and growing, and requires capital to capture at scale.”The offering is intended to further enable Intel to pursue the growth opportunities ahead,” Intel said in its statement.Intel’s Q2 results validated the offering and set the bar for Q3The capital raise arrives on the back of Intel’s strongest quarterly performance in years, reported July 23, according to Intel.Total Q2 revenue was $16.1 billion, up 25% year-over-year (YOY)DCAI segment revenue reached $6.3 billion, up 59% YOYOperating income for DCAI was $2.47 billion compared to $633 million in Q2 2025Intel Foundry revenue was $5.8 billion, up 31% YOY, with the operating loss improving to negative $2.09 billion from negative $3.17 billionClient Computing and Physical AI revenue was $8.9 billion, up 13% YOYNon-GAAP EPS of $0.42 nearly doubled the Street’s estimate of $0.22Q3 guidance calls for revenue of $15.8 billion to $16.8 billion with non-GAAP EPS of $0.38.Our Q2 results represent our strongest revenue growth in more than fifteen years.They truly do. CFO Dave Zinsner also noted that Intel is “meaningfully increasing investments in equipment, clean room space, and substrates” to support expected growth in 2026 and 2027.The $15 billion equity raise is the mechanism to fund that investment commitment.

Intel is the fifth-best-performing S&P 500 stock in 2026.Lam Yik Fei/Bloomberg via Getty Images

The strategic positioning that makes this a growth bet, not a survival moveWhat distinguishes this capital raise from dilutive offerings by companies under financial stress is the context. Intel is raising $15 billion from a position of operational strength, not necessity.Intel launched Xeon 6+, its first server-class product on Intel 18A. It introduced the OpenVINO Physical AI framework for robotics deployment at scale.Also Read: Intel’s stock split history (& prospects) explainedIt secured commercial AI deployments in physical retail, expanded ecosystem partnerships with Foxconn, Siemens, Hitachi, and Fortinet, and deepened its strategic collaboration with Google Cloud for internal AI transformation.The inference transition that AMD has also been targeting is Intel’s primary growth catalyst. As AI deployments move from training large models to running them continuously at enterprise scale, CPU demand grows proportionally. Related: Alphabet and Intel could reset the AI tradeIntel’s Xeon processor sits at the center of that demand, selected as the host CPU for Nvidia’s DGX Rubin NVL8 systems and deployed across Google Cloud, AWS, Microsoft Azure, and Tencent.The $15 billion in new equity provides the financial runway to continue that positioning across the 18A and 14A manufacturing ramps, the foundry customer commitment pipeline expected to become concrete in the second half of 2026, and the physical AI and robotics opportunity that 130-plus customer engagements are already validating.Intel is the fifth-best-performing S&P 500 stock in 2026. The equity offering is how it intends to keep earning that ranking.Related: Intel and AMD just got leverage they haven’t had in years

“Petal” debuts at No. 1 on the U.K.’s Official Hip-Hop and R&B Singles chart, earning Ariana Grande her first champion on the genre-specific list.

Enterprises running always-on AI agents keep hitting the same tradeoff. Send every task to a frontier model and the bill climbs fast. Build custom routing logic to send easy tasks to cheaper models and that becomes its own engineering project, one that has to be maintained every time a workflow changes.Nvidia is proposing a fix that touches both ends of that problem at once. The company is out on Tuesday with Nemotron 3.5 Lightning, a 30-billion-parameter open mixture-of-experts model built for high-volume, specialized agent tasks, alongside NeMo Switchyard, an open-source library that routes each step of an agent workflow to whichever model fits it best.The headline numbers: According to Nvidia, Lightning delivers up to 4x faster output than comparable models in its class, completing agentic tasks roughly 30% faster than Qwen3.6-35B at matching accuracy. Paired through Switchyard, Nvidia says the combination holds frontier-level task completion while cutting benchmark costs to roughly a third of running Opus 4.8 alone.The timing puts Nvidia in the middle of the busiest open-weight stretch the industry has seen in months. Alibaba, Moonshot, Zhipu and DeepSeek have all shipped competitive open models out of China since the spring, several landing at or near frontier performance while undercutting US labs on size or price. Meta added to that pressure by releasing its own 30-billion-parameter open agentic model, Muse Glimmer. Open weights have gone from a differentiator to table stakes in a matter of months, and Nvidia’s release lands squarely inside that shift rather than ahead of it.The pairing is the point. A model alone doesn’t solve the cost problem, and a router alone has nothing efficient to route to. Nvidia is betting that open source, applied at both the model layer and the routing layer, is what actually moves the cost needle on agentic AI, not a single cheaper model and not a smarter router bolted onto someone else’s stack.Switchyard’s real rivals aren’t other open models — they’re Not Diamond, which already powers OpenRouter’s Auto mode, and RouteLLM, the open-source framework from UC Berkeley and LMSYS. Neither ships its own model. Nvidia’s bet is that owning both sides of the decision, under one open license, is what a router-only or model-only competitor can’t match.”That is the power of a system of models, matching the right model to each step of the workflow,” Kari Briski, vice president of generative AI at Nvidia, said in a briefing.How the router actually changes the workflowModel routing isn’t a new category. OpenRouter, LiteLLM and a handful of standalone routing startups already let developers point traffic across multiple providers. Switchyard plugs into several of them rather than replacing them outright.The core problem Switchyard solves is that the right model changes as an agent moves through a task. An agent’s state shifts as tools return results, errors show up, or a step turns out to be routine rather than complex, and a fixed model choice can’t adapt to any of that.Briski described routing strategies that respond to that shifting state rather than a static task category.”It has many types of routing strategies,” Briski said. “You can have a random router, which is not that great, or you can have an agent state route or a classifier route. Depending on your routing strategy, it wants to choose the best model. In some cases you want to go with a model like Lightning for really efficient tasks, and the router will actually choose Lightning if it’s set up in your pool of models.”Cost enters the routing decision directly, not as an afterthought. In response to a question from VentureBeat, Briski said Switchyard can evaluate model verbosity, meaning how many tokens a given model tends to produce for a task, and use that prediction to steer work toward the cheaper option before the call is made.The part that keeps this from becoming its own integration project is where Switchyard sits. Nvidia split its partners into two groups: agent frameworks that call Switchyard directly, including Cognition, LangChain and Nous Research, and LLM gateways that have built Switchyard support into their own products, including Kong, LiteLLM and OpenRouter. Kong ships Switchyard natively inside Kong AI Gateway. Briski pointed to that same list of gateway partners when describing how the library fits into the existing routing ecosystem.”We are an ecosystem lover, and we want to make sure that we are integrated,” Briski said. “We’ve partnered with OpenRouter, LiteLLM and Kong, and they’ve already integrated our routing algorithm, so you can pick it up right where you’re already using the best tools.”Nvidia shared results from nine companies testing Switchyard, several with specific figures attached. LangChain reported a 74% cost reduction across 145 multi-turn Deep Agents tasks by routing just 7% of calls to a frontier model, at a 6% accuracy tradeoff. Ramp said it matched a frontier model’s performance on Ramp SWE-Bench while cutting costs 58% and runtime 33%. Cognition integrated Switchyard’s staged router into Devin Desktop for internal use and reported near-frontier performance on FrontierCode Main while cutting mean cost 28% relative to routing everything to a single frontier model.Lightning’s architecture and performance gainsNemotron 3.5 Lightning is a standalone open model in its own right, built for high-volume, specialized agent tasks rather than general-purpose use.It extends the hybrid Mamba-Transformer, latent mixture-of-experts architecture Nvidia introduced with the Nemotron 3 family in December 2025, the same line behind Nemotron 3 Super, which Nvidia uses as Lightning’s own baseline in its post-training comparisons. Positioned within a routing setup like Switchyard, it’s built to sit at the fast, cheap end of the decision rather than the frontier end, but it runs and ships independent of any router.According to the Artificial Analysis Intelligence Index, a general capability benchmark spanning nine evaluations, Lightning scores 24, tied with gpt-oss-120b and behind Nemotron 3 Super, Gemma 4 31B, Claude 4.5 Haiku and Mistral Medium 3.5, all at 30. Lightning isn’t a general-intelligence leader in its size class, and Nvidia isn’t claiming it is.The actual claim is narrower: according to PinchBench data supplied by Nvidia, Lightning matches Qwen3.6-35B’s accuracy roughly 30% faster and beats Gemma 4 26B’s accuracy at a similar completion time on PinchBench, a real-world agent task benchmark spanning coding, research and file management. That’s a speed-to-accuracy tradeoff, not a capability win.Post-training is where Nvidia says the bigger gains show up. The company shared before-and-after figures from four early-access partners: CrowdStrike’s malicious-content recall against a Nemotron 3 Super baseline, CodeRabbit’s coding router against a GPT 5.4 Nano baseline, Harvey and Trajectory’s legal task completion against an Opus 4.6 baseline, and Lila Sciences’ energy simulation work against an Opus 4.8 baseline. CodeRabbit’s case is the most specific: Nvidia says the standard NeMo Auto model recipe, trained for one epoch, built into a working router agent for $85 in about two hours.What this means for enterprisesThere is no shortage of competitive offerings in the growing market for open models. The new Nemotron Lightning release will be yet another option for organizations to consider.On the model side, Lightning’s own benchmark chart picks Qwen3.6-35B as its direct comparison point. Asked by VentureBeat directly how Lightning compares to Chinese models more broadly, Briski didn’t offer a head-to-head benchmark, pointing instead to openness and customizability as the differentiator.”Our value proposition is not just open and it’s very customizable,” Briski said.For enterprises building agentic infrastructure, three trends stand out:The routing decision is becoming dynamic instead of static. Enterprises that built agent pipelines around a single default model are being pushed toward per-step routing based on live signals like agent state and token cost, not a fixed assignment set at design time.Open source is now a cost lever at two layers, not one. Pairing an open model with an open router a vendor controls end to end is a newer argument than cheaper weights alone, and worth watching for whether other labs follow the same pattern.The competitive question shifts from best model to best system. As routing libraries mature, the differentiator moves from which model an enterprise defaults to, toward how well its routing layer matches models to tasks in production, a harder thing to benchmark and a harder thing to market.