World

Baidu’s AI Leap: How New Chips and Supernodes Challenge US Tech Dominance

Baidu unveiled new AI chips & supernodes to boost China‘s AI, aiming for self-sufficiency amidst U.S. restrictions.

November 13, 2025, 10:15 AM

A detailed, close-up view of the Baidu logo and name in Chinese characters on a digital screen. By PJ McDonnell / Shutterstock.com.

Executive Summary

Baidu unveiled new AI semiconductors, the M100 and M300 chips, and “supernode” computing products, Tianchi 256 and 512, at its annual technology conference.

These new offerings underscore China’s drive for domestic self-sufficiency in computing power amidst escalating U.S. restrictions on advanced AI chip exports.

The M100 chip is slated for early 2026 release, the M300 for early 2027, and the Tianchi supernodes for 2026, aiming to provide powerful, low-cost, and indigenously controlled AI capabilities.

The Story So Far

China is intensely focused on achieving domestic self-sufficiency in computing power, driven by escalating U.S. restrictions on advanced AI chip exports to Chinese firms. This strategic push aims to provide Chinese companies with powerful, low-cost, and indigenously controlled AI capabilities, thereby insulating its technology sector from external vulnerabilities and ensuring national technological independence.

Why This Matters

Baidu’s introduction of new AI semiconductors and supernode computing products directly addresses China’s strategic imperative for domestic self-sufficiency in computing power amidst escalating U.S. export restrictions, aiming to provide indigenous, powerful, and cost-effective AI capabilities. This move, alongside similar efforts by other Chinese tech giants, signifies a concerted push to build a robust, independent AI supply chain within China, potentially mitigating the impact of future sanctions and fostering a unique domestic AI ecosystem.

Who Thinks What?

Baidu and other Chinese tech firms are developing new AI semiconductors and advanced “supernode” computing products to bolster China’s AI infrastructure and provide powerful, low-cost, and indigenously controlled AI capabilities.

China is driving for domestic self-sufficiency in computing power and building a robust, self-reliant artificial intelligence supply chain to ensure technological independence amid escalating U.S. restrictions on advanced AI chip exports.

Baidu, the Chinese technology giant, unveiled two new artificial intelligence semiconductors and advanced “supernode” computing products in Beijing on Thursday, November 13. The announcement, made at its annual Baidu World technology conference, underscores China’s drive for domestic self-sufficiency in computing power, particularly amid escalating U.S. restrictions on advanced AI chip exports to Chinese firms. These new offerings aim to provide Chinese companies with powerful, low-cost, and indigenously controlled AI capabilities.

New AI Chips and Supernodes

The new chips, named M100 and M300, are designed to bolster China’s AI infrastructure. The M100, an inference-focused chip, is slated for release in early 2026, while the more comprehensive M300, capable of both training and inference, is expected in early 2027. Training involves building AI models from large datasets, whereas inference applies these models to process user requests and make predictions.

Baidu also introduced two “supernode” products, which leverage advanced networking to link multiple chips, aiming to overcome limitations in individual chip performance. The Tianchi 256, comprising 256 of Baidu’s P800 chips, will be available in the first half of 2026. A more powerful version utilizing 512 of those chips is scheduled for launch in the second half of the same year.

Competitive Landscape

This development follows similar moves by other Chinese tech firms in the competitive AI hardware space. Huawei, for instance, has deployed its CloudMatrix 384 product, which utilizes 384 of its Ascend 910C chips. Industry observers consider Huawei’s offering to be more powerful than Nvidia’s GB200 NVL72, a leading U.S. system-level product. Huawei also announced plans in September for more powerful supernode products in the coming years.

Enhancing AI Capabilities

Further enhancing its AI ecosystem, Baidu revealed an updated version of its Ernie large language model. The company stated that this new iteration excels not only in text processing but also demonstrates proficiency in image and video analysis. Baidu has been developing proprietary chips since 2011, reflecting a long-term strategic commitment to domestic AI hardware.

Strategic Implications

Baidu’s latest hardware and software innovations highlight China’s concerted efforts to build a robust, self-reliant artificial intelligence supply chain. These advancements are critical for the nation’s technological independence, ensuring access to advanced computing power despite external export controls. The strategic focus on domestic development is set to shape the future landscape of AI in China.