The Ascend 910C, which integrates two previous-generation 910B chips, delivers performance levels comparable to Nvidia's H100 chip, achieving 60-80% of its capabilities, and is appealing to Chinese AI developers.

Despite facing production challenges due to SMIC's technology lagging behind TSMC's, Huawei has successfully fabricated the Ascend 910C using local foundries and has begun distributing samples to various tech firms.

The Ascend 910C is produced using SMIC's 7nm process technology, although the manufacturing process has been scrutinized due to low yield rates and potential sourcing of components from TSMC.

Huawei Technologies is poised to begin mass shipments of its advanced AI chips, the Ascend 910C and the upcoming Ascend 920, to Chinese customers, a significant move in the AI hardware market amidst U.S. export restrictions.

As the tech rivalry intensifies, analysts predict that Chinese companies, led by Huawei, could dominate chip markets by 2030, benefiting from competitive pricing and local policy support.

In light of these developments, Nvidia faces significant financial challenges, with estimates indicating potential losses of up to $5.5 billion due to diminished access to the Chinese market.

This shift in the AI hardware landscape is expected to spur innovation in China, as researchers optimize local devices for AI applications, mirroring past advancements in computing.

The CloudMatrix 384 system, which incorporates the Ascend 910C, is estimated to deliver over 300 petaflops at FP16 precision, surpassing Nvidia's current offerings in memory bandwidth and capacity.

The bifurcation of the global AI ecosystem is becoming evident, with distinct technological tracks emerging for the U.S. and China, potentially limiting collaboration and interoperability.

In response to new U.S. export licensing requirements imposed on Nvidia's H20 chip in April 2025, which disrupted Nvidia's substantial business in China, local companies are actively seeking alternatives, including Huawei's 910C.

The new chip supports popular AI frameworks, facilitating a smoother transition for Chinese companies as they pivot away from reliance on Nvidia's products.