Chinese AI team wins global award for replacing Nvidia GPU with industrial chip
In a bold challenge to US giant Nvidia's dominance in artificial intelligence (AI) hardware, Chinese
researchers have trained a cutting-edge video-generation model on an off-the shelf industrial chip – outperforming high-end GPUs in both speed and efficiency.
Advertisement
Their system, FlightVGM, recorded a 30 per cent performance boost and had an energy efficiency that was 4½ times greater than
Nvidia's flagship RTX 3090 GPU – all while running on the widely available V80 FPGA chip from
Advanced Micro Devices (AMD), another leading US semiconductor firm.
The innovation earned top honours at the prestigious FPGA 2025 conference which concluded on March 1. The win marked the first time a mainland Chinese team had claimed the event's Best Paper Award, signalling a seismic shift in the global race to optimise
AI hardware
Developed by scientists from Shanghai Jiao Tong University, Tsinghua University and Beijing-based start-up Infinigence-AI, the model could redefine how industries deploy cost-effective, energy-efficient AI systems, from robotic controls to
autonomous vehicles
FPGAs or field-programmable gate arrays, are programmable semiconductor devices that allow post-manufacturing modifications to their circuitry and functionality. In contrast, conventional
chips like CPUs or central processing units, GPUs or graphics processing units, and ASICs or application-specific integrated circuits have fixed functionalities once fabricated.
Advertisement
In video-generation and general computing, FPGAs and Nvidia GPUs each have distinct advantages. FPGAs offer customisable architecture tailored to specific applications, resulting in higher energy efficiency and lower latency. Meanwhile, Nvidia GPUs, known for their massive parallel computing power, excel in processing large-scale data and handling complex computational tasks.
Building on previous research, the Chinese team developed FlightVGM, the first FPGA-trained video-generation AI model. Through innovations in data architecture and scheduling methods, FlightVGM achieved computational performance that could outpace GPUs.
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Asia Times
2 hours ago
- Asia Times
The power calculus driving Trump's tariffs
Despite dire predictions that US President Donald Trump's foreign policy, dominated by real and threatened manipulations of American tariffs and trade practices, US inflation rates and other measures of American economic vigor do not yet give cause for alarm. Indeed, at this writing US-China trade talks seem productive enough so that spokespersons for the European Union say they hope their trade talks take on a similar format. Trump's approach to tariffs has been anything but static—shifting abruptly like a spotlight sweeping across a stage. Yet beneath the political theater lies a calculated strategy with far-reaching implications. While critics assume tariffs invariably raise consumer prices, the reality is more nuanced. Trump's policies appear designed not just for economic leverage but as an extension of his foreign policy vision, particularly in Asia and the Western Hemisphere. Whether this constitutes strategic brilliance or overreach is debatable, but the mechanics of tariffs—and who ultimately bears their cost—demand closer scrutiny. The impact of a tariff hinges on market dynamics, competition and geopolitical leverage. Consider a US$100 product imported from Country X. If the US imposes a $25 tariff, the seller faces a choice: absorb the cost by cutting their price to $75 (keeping the consumer's total at $100) or pass the expense to buyers and risk losing market share. In competitive markets—like coffee from Colombia, Brazil, or Mexico—sellers often absorb tariffs to retain customers. But the calculus shifts when alternatives are scarce. A monopolist, such as OPEC in the oil markets, can dictate prices precisely because competitors lack the capacity to undercut them without facing ruin. This imbalance of power invites broader consequences: nations disadvantaged by such asymmetries may resort to political or even military retaliation, as nearly occurred during the 1970s oil crises. Tariffs also reshape local economies. A Mexican manufacturer facing US tariffs might offset losses by raising prices for domestic consumers or slashing wages. A Canadian auto supplier could lobby for government subsidies to preserve jobs while lowering export prices. Meanwhile, China's state-influenced exporters might reduce prices to maintain access to the elastic US market, repurposing tariff revenue for Chinese domestic projects. Trump's tariffs align with a modern revival of the Monroe Doctrine, which asserts US hegemony in the Western Hemisphere. Recent maneuvers, such as discouraging Chinese influence over the Panama Canal, signal that the administration views tariffs as both economic tools and geopolitical signals. The message is clear: the US will enforce its sphere of influence, and trade policy is one lever to do so. It is possible to imagine Trump's 'super big picture' plan as a compressed version (spanning three years) of the 150-year evolution of the British Empire, beginning with Mercantilism and culminating in free trade. At first, Trump treats the rest of the world as composed of client states, whose economies are tied tightly together with the 'mother ship', the dependent states all at first directed by force majeure to contribute to the greatness of the Metropolitan Authority. Later on, when the dependencies have grown to maturity, a managed form of free trade emerges, and wealth becomes more widely shared. Ultimately, outcomes will be determined by raw power—economic, military and diplomatic. While Trump's aggressive posture may yield short-term gains, inconsistency risks undermining his objectives. China, the primary challenger to this strategy, may currently perceive his actions as domestically focused rather than existential. But if tariffs become an erratic flicker rather than a steady beam, the US could squander its leverage. In an era where trade is war by other means, Trump's tariffs are less about economics than they are about reasserting American primacy. The question isn't whether the world will adapt—it's who will blink first.
South China Morning Post
4 hours ago
- South China Morning Post
AMD says new chips can top Nvidia's in booming AI chip field
Advanced Micro Devices Chief Executive Officer Lisa Su said her company's latest AI processors can challenge Nvidia chips in a market she now expects to soar past US$500 billion in the next three years. The new instalments in AMD's MI350 chip series are faster than Nvidia counterparts and represent major gains over earlier versions, Su said at a company event Thursday in San Jose, California. The MI355 chips, which started shipping earlier this month, are 35 times faster than predecessors, she said. Though AMD remains a distant second to Nvidia in AI accelerators – the chips that help develop and run artificial intelligence tools – it aims to start catching up with these new products. The stakes are higher than ever: Su previously predicted US$500 billion in market revenue by 2028, but she now sees it topping that number. 'People used to think that US$500 billion was a very large number,' she said in a briefing following her presentation. 'Now it seems well within grasp.' In February, AMD gave a forecast for its data centre business that reflected slower growth than some analysts predicted. AMD believes the new update to its MI range will restore momentum and prove it can go toe to toe with a much bigger rival. Jensen Huang, CEO of Nvidia, gestures as he speaks during the Viva Technology conference in Paris, France, June 11, 2025. Photo: Reuters AMD said that the MI355 outperforms Nvidia's B200 and GB200 products when it comes to running AI software and equals or exceeds them when creating the code. Purchasers will pay significantly less than they would versus Nvidia, AMD said.
South China Morning Post
12 hours ago
- South China Morning Post
Nvidia to stop including China in profit, revenue forecasts amid US chip export controls
Nvidia will exclude the Chinese market from its revenue and profit forecasts following the imposition of tough US restrictions on chip sales to China, CEO Jensen Huang told CNN on Thursday. Advertisement When asked whether the US will lift export controls after trade talks with China in London this week, Huang said that he was not counting on it. 'But, if it happens, then it will be a great bonus. I've told all of our investors and shareholders that, going forward, our forecasts will not include the China market,' Huang said. Nvidia did not immediately respond to a Reuters request for comment. US President Donald Trump said on Wednesday he was very happy with a trade deal that restored a fragile truce in the US-China trade war , a day after negotiators from Washington and Beijing agreed on a framework covering tariff rates. Advertisement
