News
- OpenAI published the first Jalapeño benchmarks at Hot Chips, claiming 1.5 to 1.9x more throughput per kilowatt and up to 3.6x lower latency than Nvidia's GB200 and GB300 rack systems
- The chip's purported gains are at a reported 700W power draw and center around efficiency, with Nvidia's GB300 still leading on absolute throughput per package by 20-25 percent.
- The numbers are currently self-reported, and volume production of the chip is not expected to ramp up until 2027
OpenAI took the stage at Hot Chips on August 25 with the first published performance figures for Jalapeño, the inference accelerator it co-developed with Broadcom, and the numbers are designed to be read the way it flatters the former: efficiency.
The company reported 1.5 to 1.9 times more throughput per kilowatt and 1.7 to 3.6 times lower end-to-end latency across three open models than the Nvidia rack systems it tested against.
OpenAI's Jalapeño is currently rated at 700W, versus comparable Nvidia silicon that was rated for 1200W and 1400W, a key differentiator in a benchmark that already has limited information available to researchers looking to pick a clear winner.
AI efficiency takes center stage?OpenAI's Jalapeño is an ASIC, or an Application-specific integrated circuit, which essentially means that it focuses primarily on and is great for very specific AI workloads; in this case, OpenAI's inference needs, which allow it to run multiple models with significant efficiency gains in tow.
Richard Ho, who runs OpenAI's hardware program, told reporters on a press call that the results show "a very, very significant performance advance over state-of-the-art."
OpenAI ran SemiAnalysis's public InferenceX suite on GPT-OSS 120B, DeepSeek R1 670B, and Moonshot AI's trillion-parameter Kimi K2.5, at a nominal 8,000-token input and 1,000-token output. It scored a 1.9x efficiency win vs. GB200 on its own GPT-OSS model, and 1.7x and 1.5x on DeepSeek and Moonshot's offerings against a GB300 system. Interestingly, the figures for GPT-OSS 120B running on a GB300 compared to OpenAI's Jalapeño are not published.
It is important to point out here that there is a certain degree of cherry-picking to flatter OpenAI's own results: by choosing to compare ratios on a per-kilowatt basis versus a per-chip basis, while efficiency remains consistent, it does paint Jalapeño as potentially a much more potent competitor to Nvidia's last-generation offerings than it actually is.
Nvidia's GB300 might not have an efficiency win versus Jalapeño, but it remains a much more powerful chip in all tests, at a time when Nvidia is already rolling out Vera Rubin. Curiously, OpenAI has not published any information about how its chip stacks up against an HBM4-touting Vera Rubin, which Nvidia has presented as an efficiency juggernaut.
It is important to point out that Jalapeño's results also use single-token prediction throughout, with no speculative decoding and no prefill-decode disaggregation. Nvidia's production deployments commonly use multi-token prediction, and when OpenAI pits Jalapeño against a GB300 configured that way, the peak efficiency lead drops to roughly 1.5x.
OpenAI says it built the chip to keep model state local, writing that it "designed Jalapeño to minimize data movement and communication delays." Cores and HBM are divided into slices, each core slice holding a low-latency view of its own memory, with synchronization pushed onto a dedicated collective network. Each package pairs a compute die with six HBM4 stacks for 216 GiB at 15.4 TB/s, which is less capacity than the GB300's 288GB of HBM3E but considerably more bandwidth, and roughly 50 percent more memory per watt of rated power.
Every number published currently comes from A0 silicon, also known as the initial, first-run version of the chip. A more efficient B0 stepping is already in the fab with roughly 25 percent better performance per watt, and production is scheduled to ramp gradually across 2027.
Although it may not hold a candle to Nvidia's Vera Rubin, it might not have to; Alexander Harrowell of Omdia told CNBC that "this is the biggest competitive threat to NVIDIA," noting that roughly half of AI infrastructure capital spending comes from firms that already run custom chip programs or could. Yole Group's Adrien Sanchez framed it as pressure on Nvidia's inference margins specifically, the fastest-growing part of the business.
The timing of this announcement is both awkward and potentially suspect at a time when Nvidia just agreed to backstop $105 billion in financing for OpenAI's upcoming Ohio facility, even as Richard Ho told Bloomberg that the company continues to need a lot of Nvidia's chips, with the latter being "a really good partner." This is hardly surprising, given Jalapeño's limitations as an ASIC useful mainly for inference, while Nvidia's chips are key to training new AI models.
From Nvidia's perspective, this can be seen as a blow to its plans for a high-margin inference industry where one of the two major buyers of inference hardware is not only looking for a homegrown alternative, but is already well on its way to getting the job done.
- HUR identified 140 components linked to the upgraded Iskander missile system
- American manufacturers supplied 63 of the missile's identified electronic components
- Only one-quarter of all components came from Russian domestic manufacturers
Ukraine’s Defense Intelligence (HUR) has released details about foreign-made parts it says were found inside Russia’s upgraded Iskander-1000 ballistic missile.
The disclosure said American companies, including Xilinx, part of AMD, Atmel, part of Micron Technology, DM&P Electronics, Winbond Electronics, and Analog Devices, supplied critical parts for the missile's guidance and processing systems.
The agency also identified 140 components linked to the missile and said only a limited share came from Russian manufacturers.
Foreign chips remain inside Russia’s upgraded missile systemThe disclosure points to continued foreign technology use in a weapon system developed by Moscow despite efforts to reduce outside dependence.
According to Ukrainian intelligence, 35 companies are connected to the production network supporting Russia’s Iskander-M missile family.
The agency said the updated 9M723-2 variant, also called Iskander-1000, has appeared during recent attacks and forms part of the Bora-M missile system.
The newer missile version reportedly uses a larger engine and modified launcher design, allowing it to reach distances of up to 550km.
This compares with the 390km range attributed by Ukrainian intelligence to the earlier 9M723-1 missile model.
HUR stated that Russia continues to use imported electronic parts in key missile areas despite attempts to replace foreign technology with domestic alternatives.
The agency said only about one-quarter of the identified components were produced by Russian firms.
The intelligence report lists 63 components linked to manufacturers from the United States, while Switzerland and Belarus each account for eight components.
Germany and Taiwan were each linked to four parts, while China supplied three components according to the information released by HUR.
“The most critical components—microprocessors, FPGAs, memory, digital-to-analog and analog-to-digital converters—come from American and Taiwanese manufacturers…” said Ukraine’s Defense Intelligence.
The missile’s guidance system was described as one of the areas with significant foreign dependence.
HUR claims only five of 43 parts inside a flight information processing unit connected to the missile’s laser gyroscope were Russian-made.
Sanctions face challenges as supply chains remain complexThe identified components include field programmable gate arrays, memory chips, microprocessors and conversion devices used for handling digital and analogue signals.
Such parts are common in many electronic systems, making their movement across global markets difficult to monitor completely.
Ukraine’s intelligence service has called for stronger restrictions against companies, organizations and facilities it believes are linked to Russia’s missile production network.
The agency said international cooperation would be necessary to limit access to important technologies.
However, controlling the movement of advanced electronic components remains difficult because many parts have civilian uses and can pass through several countries before reaching final users.
The presence of international suppliers does not alone establish whether companies directly supply military programs.
The report comes as Ukraine continues efforts to disrupt Russia’s missile operations through sanctions pressure and attacks against launch equipment.
Ukrainian officials have also reported strikes against Russian missile launch platforms before they could be used.
“To counter the aggressor’s plans, it is important to consolidate efforts aimed at restricting Russia’s program for the development, production, and modernization of ballistic weapons,” HUR said.
Via United24media
Tineco is showcasing its latest smart home care technology under the theme “The Future Cleans Itself,” at IFA 2026, demonstrating how automation can simplify tasks before, during and after cleaning. For years, Tineco has developed smart home care products designed to make everyday cleaning simpler, and in 2020, the brand helped pioneer the floor washer category with the launch of the FLOOR ONE S1. Since then, Tineco has continued to push the category forward, earning recognition as the world’s No. 1 household wet & dry vacuum cleaner brand for four consecutive years.
Building on that evolution and leading Tineco’s floor care showcase at IFA 2026, is the Floor One Station S9 Scientist Pro which combines powerful cleaning performance with automated features designed to simplify operation and help keep the machine ready when you need it.
One of the key conveniences of the Tineco Floor One Station S9 Scientist Pro is the base station. While some vacuum cleaners can be a chore to use and put away, the base station helps streamline both the cleaning process and the steps that follow. It provides fast charging, keeping the Tineco Floor One Station S9 Scientist Pro powered up and ready for cleaning sessions of up to 100 minutes.
The station automatically replenishes hot water and dispenses cleaning solution, reducing the need for manual refilling during cleaning. The base station’s five-liter tank holds plenty of water for frequent cleaning, even in large homes.
After cleaning, the base station works with the floor washer to flush its roller and pipes and dry them with 230°F hot air in just five minutes. An integrated comb in the head of the Tineco Floor One Station S9 Scientist Pro helps cut out hair and detangles as you clean.
All of this cuts down on the manual upkeep needed before the next cleaning session.
The Tineco Floor One Station S9 Scientist Pro brings a full arsenal of cleaning capabilities to tackle the many messes that your floor might face. Its wide brush head and 25,000 Pa of suction can tackle dry debris with ease. Powered wheels, a swiveling neck and an edge-to-edge roller make it easy to navigate and clean along edges and baseboards. The Tineco Floor One Station S9 Scientist Pro can even lie flat to get under furniture as low as 4.33 inches.
For dried, sticky or oily messes, Hydro Burst Cleaning blasts the area with high-pressure water and cuts right through the mess. As you move the floor washer forward, it operates normally, rolling up debris, but with one click it can seamlessly drop a special scraper and move in reverse to more effectively pull up liquids and wipe away water for a streak-free finish.
The Floor One Station S9 Scientist Pro is just one part of Tineco’s broader showcase at IFA 2026. The brand is also unveiling the FLOOR ONE S11 Artist Fusion Steam, FLOOR ONE S9 Artist Boost Steam and PURE ONE STATION 7 ZeroTouch, all making their global debut at the event. Beyond these floor care launches, Tineco’s IFA presence spans floor washers, carpet cleaners, vacuum cleaners and smart kitchen products, with additional new products and technologies on display across the portfolio. Together, the lineup brings Tineco’s “The Future Cleans Itself” theme to life through technology designed to make everyday home care simpler.
*Source: Euromonitor International (Shanghai) Co., Ltd.; measured in terms of the brand’s global retail sales volume (in units) of household wet & dry vacuum cleaners in 2022, 2023, 2024, and 2025. Household wet & dry vacuum cleaners are defined as household cleaners that dispense clean water (or cleaning solution) to wash hard floors and vacuum the dirty water and debris thereafter. Based on research completed in March 2026.
- Lexar PexarBolt Ultra TBT5 Dock connects to a laptop through a single Thunderbolt 5 cable
- Bandwidth defaults to 80Gbps and boosts up to 120Gbps when needed
- Runs two 8K displays or three 4K panels simultaneously
Lexar has introduced a new docking station for Mac and Windows computers, called the PexarBolt Ultra TBT5 Dock, built around Thunderbolt 5 connectivity.
The dock combines 22 ports with a 5-inch touchscreen and a rotating metal dial to control connected displays and check system status.
The dock's Thunderbolt 5 connection is rated at up to 80 Gbps by default, with a boost mode that pushes that bandwidth to 120 Gbps in one direction.
A desk hub built around high-speed linksAccording to Lexar, the dock can handle two 8K displays at 60Hz or three 4K panels running at 144Hz, depending on host hardware.
Video output comes through Thunderbolt connections alongside DisplayPort 2.1 or HDMI 2.1, although those latter outputs share the same video path.
The unit also accepts Type B CFexpress, SD 4.0 and microSD 4.0 cards, giving camera users several ways to move footage.
An internal M.2 storage slot supports PCIe Gen 4 x2, with advertised transfer rates reaching 3,600 MB/s when suitable hardware is installed.
The storage drive is not included, so buyers would need to supply their own compatible SSD for that internal bay.
Network transfers can use 10GbE, while multiple USB ports provide connections for accessories, drives, phones, tablets and other desk equipment.
On the rear, this device has two slower USB-A ports, two faster USB-A ports, five USB-C sockets, audio, Ethernet, display outputs, and charging.
The front carries the camera card readers, a headphone socket, and two USB-C connections that share their 45W power allocation, though the power delivery reaches 140W for a connected laptop.
Its compatibility covers macOS and Windows, while the dock also works with Thunderbolt and USB4 devices when their supported features allow it.
A screen and cooling add more hardwareThe 5-inch display is intended to show information about connected equipment, power use, temperatures, transfer rates, and other operating details.
A rotary control beside the screen lets users move through interface options, while taps on the display provide access to additional information.
The company says full dashboard functions require its desktop application, making the screen less independent than a simple hardware status panel.
Active cooling combines a fan, metal heat transfer parts and directed airflow, with thermal information also shown through the display.
The package includes the dock, power supply, three regional AC leads, an SSD fitting kit, a Thunderbolt 5 cable and manual.
Lexar has not listed the NVMe drive as part of the package, despite making internal storage expansion one of the dock's main features.
The company is preparing the product for Kickstarter, where the advertised early price is $479 against a regular $649 MSRP.
Disclaimer: We do not recommend or endorse any crowdfunding project. All crowdfunding campaigns carry inherent risks, including the possibility of delays, changes, or non-delivery of products. Potential backers should carefully evaluate the details and proceed at their own discretion.
- Political action committees are funneling millions of dollars into candidates and adverts in an attempt to change perspectives on AI and data centers
- Numerous states have changed regulations on data centers, making them increasingly hard to gain approval for new projects
- The US economy is hinged upon the growth of AI companies, and opposition is costing money
A new breed of AI-aligned super political action committees (SuperPAC) is spending millions of dollars in an attempt to get pro-AI data center candidates elected in this year’s midterm elections.
Data centers have become a hot political topic with the majority of Americans preferring to have a nuclear reactor in their backyard rather than an AI data center. Americans are also increasingly concerned about AI and its effects on the job market, the environment, and their own health.
But these perspectives just don’t align with AI companies, who have seen unprecedented growth in their model’s capabilities and their stock value. For that growth to continue, perspectives on AI and the infrastructure behind it need to change.
Build data centers or become ‘backwards and poor’There are a few main players behind this movement. Leading The Future (LTF) has become a key funder behind pro-AI and pro-data center candidates, but also says it is using its millions to oppose those opposed to AI - as if the midterms weren’t already confusing enough.
But a new group, Build American AI, has emerged to run pro-AI and pro-data center ads in key battleground states of Kansas, Ohio, and Wisconsin. The Republican party recently issued a call for help from AI firms over AI opposition due to fears it could cost them the Ohio senate race.
If that were to be the case, it could lead to a wider change of floating voters aligning themselves with whoever is opposed to AI data centers.
Even Nvidia is jumping on the PAC train, having recently set up NVPAC to shape policy in its own favor. Nvidia has an enormous stake in AI technologies, given that it commands an enormous share of the hardware used in most AI data centers.
The Trump administration has aligned much of America’s politics around AI. To remain a global power ahead of China, AI cannot be slowed down. To remain on top of America’s battlefields at home and abroad, AI is being integrated into every arm of the military and all aspects of law enforcement. Opposition to AI is therefore opposition to America’s future, or so Trump would want you to believe.
The US economy is currently being buoyed by the surge in growth experienced by AI companies, but the models they are creating have been labelled not only a threat to financial institutions, but also a threat to financial stability of the world.
“The only reason that communities throughout the U.S.A. should not want Data Centers is if they want to end up being backwards and poor,” Trump wrote in a Truth Social post. “If they want to be successful and rich, with far lower taxes and jobs all over the place, let Data Reign.”
Trump’s referrals to taxes and jobs raise an interesting point, because numerous states have lost out on billions of tax revenue due to data center exemptions, and numerous studies have shown AI is leading to job losses in multiple industries.
There is also a large amount of infighting within the Republican Party surrounding support for data centers. Many representatives that previously pushed for data center construction - such as Texas Gov. Greg Abbott - have done a 180.
AI could therefore go from a revolutionary technology to one of the most divisive political issues of the Trump administration and beyond.