NVIDIA Corporation (NVDA.US) financial report is approaching, Citigroup has given a slightly optimistic forward-looking judgment. The AI inference roadmap may become a new catalyst.

NVIDIA Corporation (NVDA.US) will announce its latest quarterly performance and guidance on February 25th. Citigroup has given a slightly optimistic outlook on this chip giant led by Huang Renxun, predicting that the company will release strong performance guidance. Citigroup analyst Atif Malik stated in a report to clients that his model predicts NVIDIA Corporation's revenue for the quarter ending in January to be about $67 billion, higher than the Wall Street consensus of $65.6 billion. At the same time, he expects the company's revenue guidance for the quarter ending in April to reach $73 billion, significantly higher than the market's expectation of $71.6 billion. Malik pointed out that the continuous release of the B300 product combined with the introduction of the Rubin architecture will drive NVIDIA Corporation to achieve a 34% year-on-year sales growth acceleration in the second half of 2026, significantly higher than the growth of about 27% in the first half of 2026. He believes that the focus of most investors is no longer limited to this earnings report itself, but is shifting towards the annual GTC conference scheduled for mid-March, where NVIDIA Corporation is expected to emphasize its inference roadmap, including how to use Groq's low-latency SRAM intellectual property and provide a preliminary outlook on AI-related sales prospects for 2026 to 2027. Based on the above assessment, Malik maintains a "buy" rating for NVIDIA Corporation with a target price of $270. In his analysis beyond short-term performance, Malik further points out that NVIDIA Corporation's current valuation is "attractive" from a medium to long-term perspective. With increasing visibility into the performance of 2026, the stock is expected to outperform the market in the second half of 2026. He also stated that the inference market is evolving towards a "more diversified" direction, which will provide more choices for customized model sizes and application scenarios, and also means that the use of AI accelerators will present a more diverse form. However, from a systemic perspective, he expects NVIDIA Corporation to continue to maintain a leading position in training-oriented and inference and logic inference-heavy workloads, and believes that MLPerf is still the most valuable benchmark for comparing different AI accelerators.