As 2025 winds down, leading tech stocks once again helped lead the market higher by double-digit percentages. Continued strong momentum in the sector should continue through December and into 2026, making now a good time to look at two tech stocks to see if they are worth buying before the end of the year.
Nvidia
While Nvidia (NVDA 0.64%) is starting to face increased competition, it is still the king of artificial intelligence (AI) infrastructure. The company's graphics processing units (GPUs) are the prime chips used to power AI workloads, and the company has over 90% market share in the GPU data center space.
Image source: Getty Images.
The biggest challenge to Nvidia is coming from custom AI ASICs (application-specific integrated circuits), which are pre-programmed chips that are hardwired to perform specific tasks. Because the chips are designed for a sole purpose, this makes them more energy efficient. However, they lack the flexibility of GPUs, which are adaptable and can operate within various AI frameworks. While Alphabet's tensor processing units (TPUs) have gotten a lot of praise, these custom ASICs are specifically designed to run within Google Cloud and its TensorFlow framework.
NASDAQ: NVDA
Key Data Points
ASICs are also custom chips, so each company needs to design its own, which means upfront costs are high and it takes time to develop them. Companies also need to secure advanced packaging from semiconductor manufacturers, like Taiwan Semiconductor Manufacturing (TSM +2.22%), which are largely booked up by companies like Nvidia and Apple. This makes scaling advanced ASICS difficult, given the capacity constraints the industry is facing. Nvidia's GPUs, meanwhile, are relatively readily available at scale, and the company works closely with TSMC to help secure future supply to meet increasing demand.
As such, Nvidia is still the company best positioned to benefit from the ongoing AI infrastructure boom. The company has a nice moat, as most foundational AI code has been written on its CUDA software platform and optimized for its chips. With Nvidia predicting data center capital expenditures could hit $4 trillion by 2030, it's poised to capture more than its fair share of this spending.
The stock is also attractively priced, trading at a forward price-to-earnings (P/E) ratio of under 24.5 times 2026 analyst estimates and a price/earnings-to-growth (PEG) ratio below 0.7 times (with below 1 times considered undervalued).
Taiwan Semiconductor Manufacturing
Another great stock to own heading into next year is Taiwan Semiconductor Manufacturing. The company is the largest semiconductor contract manufacturer in the world, and the leading maker of advanced chips. It is part of an oligopoly in advanced chip manufacturing with Samsung and Intel, and it's greatly benefiting from its rivals' struggles.
NYSE: TSM
Key Data Points
For chip technology to advance, chips need to become denser, meaning more transistors need to be squeezed onto a chip. This is referred to as shrinking node sizes in the chip industry, and TSMC is the only foundry that has been able to consistently produce chips at small node sizes with low defect rates at scale. Last quarter, 60% of its revenue came from chip technology at 5nm nodes or below. This has made it a key partner to chip designers, such as Nvidia, with whom it works closely on future chip designs and to increase capacity to meet growing demand.
Given the AI infrastructure boom, TSMC is aggressively building new fabs (chip manufacturing facilities) to expand capacity. With capacity tight, it also has strong pricing power, and it is expected to raise prices next year. Meanwhile, the company is set to soon introduce its newest 2nm node technology, which is projected to cost 50% more than its 3nm processing technology. The combination of increasing capacity and pricing should lead to continued strong growth for TSMC moving forward.
Meanwhile, the stock is also still attractively valued, trading at a forward P/E of 24 times. Given its growth outlook, this makes the stock a solid buy at today's levels.
