Skip to main content

Iluvatar CoreX Soars by 31.54% in Trading Debut: 'Genuine Leader' of Domestic AI Computing Power Chips Emerges

HONG KONG, January 8, 2026 - (ACN Newswire) – On January 8, Shanghai Iluvatar CoreX Semiconductor Co., Ltd. (“Iluvatar CoreX”, stock code: 9903.HK), the highly anticipated Chinese GPGPU chip design company, officially commenced trading on the Main Board of the Hong Kong Stock Exchange. The stock surged over 31.54% at the opening bell, propelling its market capitalization past the HK$ 48.37 billion threshold.

The impressive first-day performance not only validated the market's strong expectations for the substitution of domestic computing power but also highlighted the robust investment value of Iluvatar CoreX, serving as the “computing power chip backbone” for domestic large models. As a scarce GPGPU stock in Hong Kong's equity market, Iluvatar CoreX has emerged as the most definitive core investee amid the AI investment boom by virtue of its lucid business logic and solid commercialization achievements, offering an optimal timing window for investors to position themselves within the domestic high-end computing power sector.

Full-stack technological upgrades solidify the computing power foundation for domestic large models

At the pivotal juncture where AI large models transition from technical validation to scaled deployment, the technical maturity and consistency of supply of GPGPU directly determine the pace and penetration of industrial implementation. Through its fully-stack self-developed technology path and continuously evolving GPGPU architecture, Iluvatar CoreX has emerged as the most dependable “computing power chip backbone” for domestic large models, breaking the monopoly held by international giants in the high-end computing power sector.

As the China’s first company to achieve mass production of GPGPU chips for both training and inference, Iluvatar CoreX has established a product ecosystem spanning the entire lifecycle of large models: the “TG” series training chips prioritizes massive data processing and model training, addressing the high-intensity computing demands of large-scale parameter models; while the “ZK” series inference chips are optimized for deployment scenarios and address inference-specific requirements such as low latency, high throughput, and optimal power efficiency, delivering excellent performance-to-cost ratio for large language model inference tasks. Crucially, adhering to the "hardware-software integration” philosophy, Iluvatar CoreX has independently developed a software stack that ensures full compatibility with mainstream GPGPU programming ecosystem and platform, thereby significantly reducing customer migration costs.

Through three generations of GPGPU architectural iterations and progressive software stack enhancement, Iluvatar CoreX has achieved steady advancement in computing performance, ecosystem compatibility, and adaptability across diverse application scenarios, positioning itself as a genuine technology-driven enterprise rather than a “concept company” reliant on policy incentives. As domestic large models transition from technical validation to in-depth industry integration and scaled deployment, the demand for a high-performance, cost-effective, and consistently supplied domestic computing backbone is poised for explosive growth. Leveraging its first-mover advantage in mass production, proven product reliability through bulk deployment, and extensive hardware/software ecosystem compatibility, Iluvatar CoreX has become a serious option for numerous technology giants and industry leaders pursuing computing autonomy.

With a lucid business logic, Iluvatar CoreX stands as the most robust AI computing chip enterprise

Amidst the AI chip industry’s prevailing challenges of “splashing the cash with narrow profitability”, Iluvatar CoreX has charted a healthy trajectory of “technology implementation - scaled monetization - sustained growth” by virtue of its lucid business logic, establishing itself as the computing chip enterprise with the most solid capabilities for sustainable development in China.

Unlike business models that rely heavily on government subsidies, research projects, or single demonstration orders, Iluvatar CoreX rapidly penetrated the market by leveraging its products’ advantages of enhanced efficiency, strong adaptability and high portability, with its customer base directly anchored to the most authentic and sustainable demand sources within the AI industry: cloud service providers, large data centers, and leading government and enterprise clients. The core procurement logic for these clients is acquiring continuous, stable, and scalable computing to support their massive AI training and inference operations. This generates predictable bulk orders and sustained repurchase potential for Iluvatar CoreX, aligning its revenue characteristics with rigid infrastructure investments, which forms a sustainable business model closely aligned with “directly delivering computing value”. From an investment perspective, Iluvatar CoreX’s business trajectory offers greater clarity and stronger cash flow visibility.

Meanwhile, Iluvatar CoreX maintains an efficient and controlled pace in its R&D investments. Based on publicly available financial data, the “R&D expense/revenue” ratio of Iluvatar CoreX ranges between approximately 80% and 120%, representing the lowest level among comparable peers. This healthy return on investment ensures continuous technological iteration while avoiding operational pressures from reckless spending, which effectively reduces the risk of the Company being forced into strategic compromises or marginalized due to the pressure on the capital chain. Market analysis widely recognizes Iluvatar CoreX as the most likely enterprise to maintain its independent leadership position in long-term competition through its self-sustaining capabilities and technological accumulation.

Performance continues to improve with promising growth prospects

Solid business logic brings impressive performance. As of June 2025, Iluvatar CoreX had cumulatively shipped over 52,000 units of GPGPU products, served over 290 cross-sector customers, and completed over 900 deployments and applications in essential sectors including financial services, healthcare and transportation. Revenue scale has experienced explosive growth, surging from RMB189 million in 2022 to RMB540 million in 2024, representing a compound annual growth rate of 68.8%. Revenue further increased by 64.2% to RMB324 million in the first half of 2025.

What is even more noteworthy is its high-quality profit structure. In 2024, the gross margin for the Company’s GPGPU products reached 56.6%, and the overall gross margin remained at a high level of 50.1% in the first half of 2025. This data fully demonstrates that the Company’s products have gained genuine market recognition through technological premium pricing rather than relying on low-price competition, laying a solid foundation for long-term sustainable growth.

The impressive performance of Iluvatar CoreX on its first day of listing represents the market’s “vote” for its solid accumulation over the past years and its core position in the golden track of AI computing. It has demonstrated its profit-making capability through clear business logic and its spending efficiency through healthy financial indicators. As AI investment transitions from conceptual phase to performance validation, companies like Iluvatar CoreX possessing technical hard power, genuine commercial momentum, and financial stability are poised to consistently deliver growth value amid the dual waves of AI technological revolution and domestic substitution.

]]>



Copyright 2026 ACN Newswire . All rights reserved.

Recent Quotes

View More
Symbol Price Change (%)
AMZN  247.19
+0.91 (0.37%)
AAPL  259.70
+0.66 (0.25%)
AMD  203.81
-0.87 (-0.42%)
BAC  55.91
-0.27 (-0.49%)
GOOG  330.24
+4.23 (1.30%)
META  652.32
+6.26 (0.97%)
MSFT  479.61
+1.50 (0.31%)
NVDA  185.38
+0.34 (0.18%)
ORCL  199.51
+10.36 (5.48%)
TSLA  445.44
+9.64 (2.21%)
Stock Quote API & Stock News API supplied by www.cloudquote.io
Quotes delayed at least 20 minutes.
By accessing this page, you agree to the Privacy Policy and Terms Of Service.