(Sharecast News) - Shares in American semiconductor giant Nvidia jumped in after-hours trading on Wednesday after the firm's hotly anticipated third-quarter figures blew past analysts' expectations, easing recent concerns that a potential bubble is brewing in the US-listed AI sector.

Revenues reached a record $57.0bn in the third quarter, up 22% from the second quarter and 62% higher than a year ago. This was comfortably ahead of the consensus estimate of $54.7bn.

Looking ahead, Nvidia also raised its guidance for the fourth quarter, targeting revenues of $65bn, ahead of the $62.1bn currently pencilled in by the market.

The stock, which has felt the brunt of the recent market-wide sell-off, was trading 4.2% higher at $194.42 shortly after the closing bell, building on a 2.9% gain during opening hours.

Prior to Wednesday, the stock had dropped over 12% since the start of November. Markets view Nvidia - the world's largest listed company with a market cap of over $4.5trn - as a figurehead in the AI industry troubled by eye-watering valuations that many analysts fear are overstretched.

"Nvidia bears the weight of the world, but like Atlas, it's standing firm under that towering mountain of expectations. Third-quarter results delivered the goods and then some," said Matt Britzman, senior equity analyst at Hargreaves Lansdown.

"There are certainly pockets of the AI space where valuations needed to take a breather, but Nvidia is not in that camp. In fact, while shares have performed well this year, the valuation has gotten more attractive as earnings growth has raced ahead."

Nvidia said that gross margins improved to 73.4% in the third quarter, from 72.4% three months earlier, helping to drive net income 21% higher quarter-on-quarter to $31.9bn, up 65% year-on-year. As a result, diluted earnings per share surged 67% on last year to $1.30, beating the $1.24 expected by the market.

Chief executive and founder Jensen Huang said that sales of Blackwell chips - Nvidia's new family of high-performance graphics processing units (GPUs) used in AI and accelerated computing - were "off the charts", while its cloud GPUs were completely sold out.

Huang said just last month that Nvidia had a $500bn pipeline of AI chip orders following some huge deals announced recently.

"Compute demand keeps accelerating and compounding across training and inference - each growing exponentially. We've entered the virtuous cycle of AI. The AI ecosystem is scaling fast - with more new foundation model makers, more AI startups, across more industries, and in more countries. AI is going everywhere, doing everything, all at once," Huang said in a statement to shareholder on Wednesday.