Nvidia (NASDAQ:NVDA) has ceased production of its H200 artificial intelligence chips specifically designed for the Chinese market, according to a report from the Financial Times.
The decision marks a significant strategic pivot as the company grapples with a complex web of U.S. export controls and cool reception from Beijing.
The Santa Clara-based chip giant has reportedly directed its primary manufacturing partner, Taiwan Semiconductor Manufacturing (TSMC), to reallocate capacity previously reserved for the H200 toward its next-generation "Vera Rubin" hardware.
The move suggests Nvidia is prioritizing its latest high-performance architecture—unveiled at CES 2026—over a China-bound product line that has struggled to gain commercial traction.
The halt comes despite the Trump administration's formal "green light" in January to allow H200 sales to select Chinese customers under strict conditions, including a 25% "duty" paid to the U.S. government.
However, as of late February, U.S. Commerce Department officials confirmed that not a single H200 chip had been sold to Chinese buyers.
Shipments remained stalled due to rigorous third-party verification requirements and a 50% volume cap relative to U.S. domestic sales.
Analysts suggest the production shift reflects a "limbo" state in the Chinese market.
While Nvidia recently secured licenses for "small amounts" of H200 exports, the company told investors in February that it had yet to generate revenue from these sales and omitted China data center revenue from its Q1 2026 outlook.