Aolani, a Singapore-founded neocloud powering AI growth, announced the launch of the Aolani Token Factory, a managed inference platform that enables organisations to deploy and scale AI models on a pay-per-token basis without provisioning or managing the underlying GPU infrastructure. The launch of the token factory will make Aolani the first Singapore-founded neocloud to offer production-grade, managed inference at scale.
As global AI companies expand their operations in Singapore, and companies around the world invest in AI to drive real business outcomes, the demand for production-grade inference infrastructure is quickly accelerating. Aolani Token Factory is designed to close the accessibility gap, giving AI-native companies and companies a highly compliant and high-performance path from AI experimentation to production-scale deployment.
The Aolani Token Factory will be offered as a per-token metering model, where customers can pre-purchase credits and pay based on token consumption, rather than investing in capital-intensive GPU infrastructure. Aolani manages the full inference stack, including GPU capacity allocation, model serving, orchestration, scheduling, and workload optimisation, which allows customers to scale consumption without continuously provisioning additional infrastructure.
The platform supports leading open-source models at launch, including DeepSeek, GLM, Kimi, and Qwen. Aolani plans to further expand the model catalogue over time based on customer demand and availability. Customers can also deploy their own models through OpenAI-compatible APIs. Dedicated capacity and data isolation options are available for enterprise customers with strict compliance and data residency requirements.
Sea Xu, Applied AI Research Lead at Aolani, said: "The Aolani Token Factory is built on a high-performance inference stack that supports the most in-demand open-source model families. We designed the platform for fast model adaptation and deployment, so our customers can get access quickly as new models emerge. As Southeast Asia's AI ecosystem evolves and grows rapidly, it is our goal to ensure that the infrastructure serving it keeps pace."
