NVIDIA posts $81.6bn Q1 revenue and launches $80bn buyback
NVIDIA has reported first-quarter revenue of $81.6bn, ahead of the $78.9bn consensus estimate, as the chipmaker continued to benefit from surging demand for AI infrastructure hardware. Adjusted earnings per share came in at $1.87, above the $1.76 expected by analysts. The company simultaneously announced an $80bn share buyback programme, a signal that management considers the balance sheet strong enough to begin returning capital at scale alongside growth investment.
Forward guidance was equally striking. NVIDIA guided for Q2 revenue of approximately $91bn, plus or minus two per cent, against an analyst consensus of around $86.8bn. The top end of that range would imply close to 100% year-on-year revenue growth in the current quarter.
The free cash flow story
Free cash flow for the quarter reached nearly $49bn, with $20bn returned to shareholders in the period, described by Hargreaves Lansdown senior equity analyst Matt Britzman as "a record". Britzman, whose employer is the UK's largest retail investment platform, noted that by the firm's projections, NVIDIA's free cash flow could approach $200bn for the full financial year.
"The buyback era has arrived at NVIDIA, and it is arriving with an $80bn bang," Britzman said, adding that revenue and earnings had exceeded "already high expectations". He attributed the durability of NVIDIA's market position to its breadth across chips, networking, software and full data-centre systems, arguing that this integration makes it difficult for rivals or in-house alternatives to match at scale. Britzman, who disclosed a personal shareholding in NVIDIA, maintained the view that AI infrastructure demand and customer spending plans support earnings power "remaining stronger for longer".
Market context and competitive read-across
NVIDIA's results land at a moment when enterprise capital expenditure on AI infrastructure remains elevated across hyperscalers. Microsoft, Alphabet and Meta have all signalled multi-year AI build-outs that lean heavily on NVIDIA's H-series and Blackwell GPU architectures. The announcement of an $80bn buyback is notable precisely because it suggests NVIDIA believes it can sustain both aggressive reinvestment and shareholder returns simultaneously, a combination that has historically been rare among capital-intensive hardware vendors.
The competitive landscape, however, is evolving. AMD continues to develop its Instinct GPU line for AI workloads, Intel is investing in Gaudi accelerators, and a number of hyperscalers are progressing custom silicon programmes, including Google's TPUs and Amazon's Trainium and Inferentia chips. The pace at which these alternatives can close the software-ecosystem gap, particularly around CUDA, remains the central question for any assessment of NVIDIA's long-term pricing power.
Separately, export-control policy continues to cast a shadow over NVIDIA's addressable market. US Bureau of Industry and Security restrictions on high-end chip exports to China have required the company to develop downgraded variants for that market, and further tightening remains a live risk as US-China technology tensions persist. Any escalation could affect the Q2 guidance range.
Near-term investor attention will focus on Blackwell architecture ramp rates, data-centre customer order backlogs, and whether NVIDIA's software stack, including the CUDA ecosystem and enterprise AI libraries, continues to deepen its hold on the AI development workflow as rivals compete on price and performance.