Cerebras Systems (CBRS) is expected to benefit from rising demand for low-latency artificial intelligence inference, Morgan Stanley said in a client note emailed Monday.
Latency is the time it takes to generate AI tokens, or tiny units of data, according to information posted on Nvidia's (NVDA) website. Low latency is crucial for real-time AI applications as it enhances user experience.
However, low latency often increases costs due to the need for more powerful hardware and real-time processing capabilities.
"As AI workloads become increasingly reasoning-intensive, demand for fast, low-latency inference is growing rapidly," Morgan Stanley said. "Supported by a large contracted backlog and a 750 (megawatt) committed capacity agreement, we believe Cerebras is well positioned to capture this opportunity."
Shares of the chip designer, which went public last month, soared 18% intraday Monday.
The brokerage initiated Cerebras' coverage with an overweight rating and a $250 price target.
"This is a unique chance to invest in an AI processor company with a first-mover advantage against (Nvidia), and offers substantial upside as the category evolves," according to the report.
The low-latency segment could account for 10% or more of inference hardware sales over the next few years, the brokerage said.
Contracted revenue will bring in $6 billion in revenue, or 1% market share of overall processors, by calendar year 2028, Morgan Stanley said, pointing to upside potential.
However, competition in the category is expected to intensify over time.
"Nvidia's acquisition of the assets of Groq -- an AI processing company specializing in a low latency approach -- is worth watching," Morgan Stanley said.
Price: $235.30, Change: $+34.29, Percent Change: +17.06%
