Last Updated -
July 25, 2026
Explore the business model, global strategy, and market performance including insights into its position in China.
Biren Technology, formally Shanghai Biren Technology Co., Ltd., was founded in 2019 and is headquartered in Shanghai, China. It is a fabless semiconductor company, meaning it designs chips and related systems while relying on outside foundries and contract manufacturers for production. Biren focuses on general-purpose graphics processing units, or GPGPUs, used for artificial intelligence training, AI inference, and high-performance computing. Its main offerings include GPGPU chips, accelerator cards, servers, intelligent-computing clusters, and supporting software.
The company has developed from a pre-revenue chip developer into an early-commercialization supplier of domestic AI-compute infrastructure in China. It began generating revenue from specialist technology products in 2023, completed volume production and commercialization of BR106 and BR166 products in 2025, and delivered multi-thousand-card intelligent-computing cluster projects for national intelligent-computing centers, telecom operators, AI data centers, large-model companies, and enterprise customers. Biren’s strategy is built around an integrated chip, system, software, and application approach, with a focus on serving China’s demand for local AI accelerator supply.
Biren listed H shares on the Hong Kong Stock Exchange on January 2, 2026 under stock code 06082. No Q1 2026 financial results were available in the company’s public filings as of July 22, 2026, so the latest formal figures are for FY2025. Revenue rose 207.2% to RMB1.03 billion, gross profit reached RMB557.0 million, and gross margin was 53.8%, while R&D expenses of RMB1.48 billion and an adjusted loss of RMB873.8 million show the scale of ongoing investment. The company ended 2025 with RMB2.90 billion in cash resources and disclosed RMB5.63 billion of net IPO proceeds to fund R&D, capacity expansion, and commercial deployment.
Biren Technology is a fabless AI-chip company built around general-purpose GPU technology for artificial intelligence and high-performance computing. It designs chips, accelerator cards, servers, cluster systems, and software, then relies on suppliers and contract manufacturers for production rather than owning semiconductor fabrication plants. Its commercial model is still young, with specialist technology revenue starting in 2023 and a public listing in Hong Kong in January 2026.
The company makes money mainly from intelligent-computing solutions sold to Chinese AI infrastructure customers. These solutions combine Biren GPGPU chips with accelerator cards, servers, cluster deployments, and software for AI training, inference, and HPC workloads. Biren also reports revenue from support and extended warranty services, agent fees recognized on a net basis, and rental income from intelligent-computing clusters.
FY2025 revenue was RMB1.03 billion, up 207.2% from 2024, showing that commercialization accelerated after a low base. Gross profit was RMB557.0 million and gross margin was 53.8%. The company remained in investment mode, with RMB1.48 billion of R&D expense and an adjusted loss of RMB873.8 million. No Q1 2026 financial statements were available in company filings as of July 22, 2026, so the latest formal financial period is FY2025.
Biren’s main competitive advantage is its position as a domestic Chinese supplier in a market where access to advanced foreign GPUs is constrained by export controls. Its filings highlight commercial deployment in thousand-chip clusters and capabilities aimed at clusters with more than ten thousand GPGPU chips. This matters because large AI model training and large-scale inference depend on chip performance, networking, reliability, cluster management, and software compatibility rather than silicon performance alone.
The company’s market position is promising but still early. According to prospectus data, China’s intelligent-computing-chip market grew from US$1.7 billion in 2020 to US$30.1 billion in 2024 and is forecast to reach US$201.2 billion in 2029. Biren’s disclosed 2024 revenue share was only 0.16% of China’s intelligent-computing-chip market and 0.20% of China’s GPGPU market. That small share shows large room for growth, while also confirming that Biren remains a challenger rather than a scaled leader.
The competitive benchmark is Nvidia. China’s GPGPU market was highly concentrated in 2024, with the top two players holding 98.0% share and a U.S. GPU company accounting for 97.6%. Biren’s opportunity comes from Chinese customers seeking domestic supply, especially in AI data centers, telecom infrastructure, cloud and internet platforms, large language model companies, government-linked compute centers, and enterprises. Its challenge is that Nvidia’s CUDA software ecosystem, product cadence, and installed base remain difficult to match.
Domestic competition is also intense. Biren competes with Huawei’s Ascend ecosystem and Chinese AI-chip peers including Moore Threads, Enflame, MetaX, Iluvatar CoreX, and Cambricon. Compared with Moore Threads, another Chinese GPU company, Biren is more clearly positioned around data-center GPGPU clusters for AI training, inference, and HPC, while both companies benefit from China’s localization drive and face the same need to build software ecosystems and reliable supply chains.
China is central to Biren’s business model. The company is headquartered in Shanghai, sells primarily to Chinese customers, and depends on Chinese AI infrastructure demand. Customer concentration remains a key investor issue. During recent disclosed periods, the top five customers accounted for more than 90% of revenue, including 97.9% in the first half of 2025. This concentration supports rapid scaling when major projects are won, but it also increases revenue volatility if deployments are delayed, reduced, or lost.
Biren entered 2026 with stronger funding after raising RMB5.63 billion of net proceeds from its H-share listing. The proceeds are intended for R&D, capacity expansion, and commercial deployment. At the end of 2025, total cash resources were RMB2.90 billion, while inventories were RMB948.6 million as the company built supply to meet demand and improve resilience. For investors, the business model offers exposure to China’s AI-compute localization cycle, balanced by high R&D spending, export-control risk, customer concentration, and the need to prove that its hardware and software stack scales against Nvidia and leading domestic rivals.
China is Biren Technology’s core market. The Shanghai-based fabless GPGPU company sells domestic AI-compute products to Chinese intelligent-computing centers, telecom operators, commercial AI data centers, AI and large-model companies, and enterprises. FY2025 revenue reached RMB1,034.6 million, up 207.2%, with gross margin of 53.8%, but R&D spending of RMB1,476.1 million kept the company loss-making. No Q1 2026 financial results were available as of July 2026. Biren’s local strategy is domestic substitution across chip, accelerator card, server, cluster, and software layers. In 2025 it completed volume production and commercialization of BR106 and BR166, delivered multiple multi-thousand-card cluster projects, and raised RMB5,631.4 million from its January 2026 Hong Kong listing. Its 2024 China share was still small at 0.16% of intelligent-computing chips and 0.20% of GPGPU revenue. Main competitors include Nvidia, Huawei Ascend, Moore Threads, Enflame, MetaX, Iluvatar CoreX, and Cambricon.
Biren Technology entered 2026 at a turning point: it had moved from early commercialization into larger customer deployments, while still carrying the cost structure of a company investing ahead of scale. FY2025 revenue rose 207.2% to RMB1,034.6 million, and gross profit rose 210.8% to RMB557.0 million, with gross margin of 53.8%. The company also reported an adjusted loss of RMB873.8 million, and R&D expenses of RMB1,476.1 million exceeded revenue, showing that growth is still being funded by heavy investment rather than self-sustaining earnings. No Q1 2026 financial results had been published, so the latest formal financial base remains FY2025.
Product expansion is likely to remain focused on full-stack intelligent-computing solutions rather than chips alone. Accelerator cards, servers, cluster rentals, support services, and vertical solutions in areas such as AIGC, AI coding, fintech, smart manufacturing, education, and government services extend Biren’s revenue opportunities beyond component sales.
Geographic expansion is mainly a China story. The company’s customers and growth thesis are tied to Chinese AI infrastructure, telecom, data-center, government, cloud, and enterprise demand. International expansion is constrained by export-control exposure and the highly concentrated global GPU market.
Biren’s future outlook depends on whether it converts China’s AI-compute localization demand into repeat deployments while narrowing software and ecosystem gaps. The IPO strengthens liquidity, and FY2025 showed clear revenue traction, but investors should watch cash burn, customer concentration, supply-chain resilience, and evidence that cluster-level deployments lead to durable gross profit and operating leverage.
This Company Profile was written by Dominik Diemer
Disclaimer:
All logos are the property of their respective owners and are used for illustrative purposes only.
This site is not affiliated with or endorsed by any of the featured companies.
No Financial Advice.
