무슨 일이 일어났나요?
Bitdeer AI, a subsidiary of Bitdeer Technologies Group (Nasdaq: BTDR), disclosed that it has secured off‑take commitments covering more than 70% of the 21.7 MW capacity at its A201 data center in Johor Bahru, Malaysia. The contracts are expected to generate over $1.7 billion in revenue over a five‑year period, bringing the company’s total expected revenue backlog to roughly $2.9 billion. The facility, a liquid‑cooled, rack‑scale deployment built for NVIDIA GB300 NVL72 GPUs (with some space for the upcoming Vera Rubin platform), is slated for energization in the first quarter of 2027. Bitdeer AI has already procured more than 100 racks of the NVIDIA systems ahead of energization and is targeting up to 350 MW of AI‑cloud capacity by Q1 2028 across its Malaysia campus.
Bitdeer AI announced that more than 70% of the 21.7 MW capacity at its A201 data center in Johor Bahru, Malaysia, is already covered by customer off‑take commitments. These contracts are projected to deliver over $1.7 billion in revenue across a five‑year horizon, raising the company’s total expected revenue backlog to about $2.9 billion.
The A201 facility is a liquid‑cooled, purpose‑built data center designed for rack‑scale deployment of NVIDIA’s GB300 NVL72 GPUs, with a portion of the rack space allocated for the upcoming Vera Rubin platform. Bitdeer AI has procured more than 100 racks of these systems from an unaffiliated third‑party supplier ahead of the facility’s energization, which is scheduled for Q1 2027.
Beyond A201, Bitdeer AI aims to reach up to 350 MW of AI‑cloud capacity by the first quarter of 2028, with the A201 and recently announced A202 sites together accounting for 86.8 MW. The company’s active for AI‑cloud capacity exceeds $10 billion, and it expects this figure to grow as more customers commit to its services.
왜 중요한가요?
The announcement signals a major infusion of committed revenue into the Southeast Asian AI‑infrastructure market, where power availability and proximity to fast‑growing enterprise AI demand are strategic advantages. A $1.7 billion contract backlog demonstrates strong customer demand for high‑density GPU compute, which could accelerate AI model training and workloads for regional firms. By financing the build‑out primarily through customer pre‑payments and cash flow, Bitdeer AI reduces reliance on external capital, potentially lowering financing costs and speeding deployment. The procurement of NVIDIA GB300 NVL72 racks also underscores the industry’s shift toward next‑generation GPU platforms, influencing hardware supply chains and competitive dynamics among cloud providers.
The $1.7 billion contract backlog provides a concrete revenue foundation for Bitdeer AI, reducing financing risk and enabling the company to fund its expansion primarily through customer pre‑payments and operating cash flow. This financing model can lower the cost of capital compared with equity‑heavy builds, potentially translating into more competitive pricing for end‑users.
Securing a majority of capacity ahead of energization demonstrates strong market confidence in Bitdeer AI’s ability to deliver high‑performance GPU compute. This confidence may encourage other AI‑focused enterprises in the region to consider Bitdeer’s services for training large language models, , and other compute‑intensive workloads.
The procurement of NVIDIA GB300 NVL72 racks aligns Bitdeer AI with the latest GPU technology, which offers higher throughput and efficiency than previous generations. This hardware advantage can improve the cost‑per‑compute metric for customers, making AI projects more economically viable.
대화형 메커니즘: 실제로 작동하는 방식
이 개발의 이면에 있는 기본 기술을 대화식으로 살펴보세요.
Which component of an AI application is the machine-learning model itself?
다음에 무엇을 볼 것인가
Investors and industry observers should monitor the actual energization timeline for the A201 facility and any subsequent off‑take agreements for the planned A202 site, which together will total 86.8 MW. The company’s ability to meet its 350 MW capacity target by early 2028, as well as the evolution of its $10 billion , will be key indicators of market traction. Additionally, the terms of the off‑take contracts—such as pricing structures, duration, and any performance clauses—will affect the profitability and scalability of Bitdeer AI’s AI‑cloud services.
The actual energization date of the A201 facility and any potential delays due to power availability, equipment delivery, or regulatory approvals.
The signing of additional off‑take contracts for the A202 site and the overall progress toward the 350 MW capacity target by early 2028.
Details of the contract terms, such as pricing mechanisms, duration, and any performance guarantees, which will affect the long‑term profitability of Bitdeer AI’s AI‑cloud services.
Market response to Bitdeer AI’s expansion, including any competitive moves by other cloud providers targeting the Southeast Asian AI compute market.