SAN JOSE, California, March 17 (Reuters) - Nvidia ( NVDA ) is preparing a version of its Groq artificial-intelligence chips that can be sold to the Chinese market, two sources familiar with the matter told Reuters on Tuesday.

Nvidia ( NVDA ) licensed technology from Groq, an AI chip startup, late last year in a $17 billion deal and showed a new lineup of products based around its chips at its annual developer conference in San Jose, California, this week.

The move to develop a version of the chips for the Chinese market comes as Nvidia ( NVDA ) CEO Jensen Huang said that the company has restarted production of its H200 chips, the predecessor to its current flagship chip, after obtaining export licenses from U.S. President Donald Trump's administration and purchase orders from Chinese customers.

Nvidia ( NVDA ) plans to tap Groq's chips for what is known as inference, where AI systems answer questions, write code or carry out tasks for users. In the products Nvidia ( NVDA ) showed this week, the company plans to use its forthcoming Vera Rubin chips, which cannot be sold in China, in combination with the Groq chips.

While Nvidia ( NVDA ) dominates the market for training AI systems, it faces much more competition in the inference market. Several major Chinese firms, including AI heavyweights such as Baidu, already produce their own inference chips.

Huang underscored the change in his speech at the developer conference on Monday. "The inference inflection has arrived," he said. "And demand just keeps on going up," he added, saying the revenue opportunity for Nvidia AI chips may reach at least $1 trillion through 2027. 

The chips being readied for China are not downgraded versions or made specifically for the Chinese market, one of the sources told Reuters. But the new variant can be adapted to work with other systems, the source said, adding that the Groq chip is expected to be available in May.

Nvidia ( NVDA ) did not immediately respond to a request for comment.