[For more diverse corporate information on the startup featured in this article, please visit the "Data Lab" of the Unicorn Factory big data platform .]
U.S.-based GMI Cloud, which provides high-performance GPU (graphics processing unit) infrastructure and AI (artificial intelligence) inference services, announced on the 1st that it has secured $668 million (approximately 910 billion won) in Series B funding.
Founded in 2021, GMI Cloud offers GPU clusters, inference services, and agent infrastructure as a single cloud platform. It is one of seven companies designated by NVIDIA as "Reference Platform Cloud Partners," supporting enterprises and developers in building, deploying, and scaling AI services.
This round was led by ARCHIV, an AI and robotics-focused investment firm based in San Francisco, U.S., with participation from NVIDIA. In Korea, KB Investment, DSC Investment, Kyobo Life Insurance, and KT invested. Investors from the Asia-Pacific region, including Trend Micro of Taiwan, also participated.
Of the total amount raised, $223 million (approximately 303.4 billion won) was allocated to equity investment. The remaining $445 million (approximately 605.3 billion won) was provided by CTBC of Taiwan in the form of credit extension.
Key clients of GMI Cloud include △Fireworks, △Hicksfield, △Nus Research, △OpenRouter, △Reflection, △Cartesia, △Trend Micro, and △Utopia Studio.
Contract-based ARR (Annual Recurring Revenue) has increased by more than nine times compared to the end of last year. The ARR for services currently in operation also grew by more than 4.5 times over the same period. The inference platform processes approximately 4 trillion tokens per week.
GMI Cloud will use this investment to expand GPU supply capabilities in the Asia-Pacific region, including the U.S. and Taiwan. Centered on the Taiwan AI Factory, announced as its first Asian hub, and the sovereign AI concept for Japan unveiled earlier this year, the company will scale up its infrastructure while also investing in enhancing inference services and securing key talent.
Alex Yeh, CEO of GMI Cloud, stated, "In AI infrastructure, delivery deadlines are a promise to customers, who base their product launches, hiring, and revenue plans on them," adding, "We will build reliable computing infrastructure across multiple continents and grow the product ecosystem on top of it."
Chenyu Zhao, co-founder of Fireworks, said, "Computing resources that are not secured in time cannot be utilized," noting that "GMI Cloud is a core partner with outstanding capabilities and high reliability among those supplying NVIDIA GB200 and GB300 NVL72 systems."
[MoneyToday startup media platform 'Unicorn Factory']
in FUTURE
