ExcelScale launches AI infrastructure platform in Malaysia

ExcelScale has launched an AI infrastructure platform in Malaysia, providing access to more than 100 AI models through a single OpenAI-compatible API.

ExcelScale has launched an AI infrastructure platform in Malaysia that gives organizations access to more than 100 AI models through a single OpenAI-compatible API.

The platform provides a common interface for organizations using models from different AI providers. Working directly with multiple providers can require separate accounts, APIs, billing systems, and governance controls.

Todd Abraham, general manager of ExcelScale, said the company chose Malaysia because of opportunities it sees in the local market.

"We're in Malaysia because we see the opportunity in the Malaysian market," Abraham said. "We want to contribute to helping small businesses, developers, startups, and enterprise customers leverage the capabilities of AI in a controlled way and access that technology as they need it."

Managing multiple AI providers

ExcelScale's standardized API allows development teams to connect to different models without creating a separate integration for each provider. The platform supports models for text generation, visual recognition, voice synthesis, video generation, and data analysis.

Developers can select models for different workloads while retaining the same API interface. Adding or switching supported models does not require teams to build a separate connection to each provider.

"There's no standard AI interface to connect people to that ecosystem," Abraham said. "Each vendor requires a different integration."

"Going through that process means, right now, learning maybe a new SDK, rewriting an integration, retesting, and redeploying," Abraham said.

Abraham said AI adoption has moved from experimentation into production, while the infrastructure used to manage it has not moved at the same pace.

"The gap is not capability. There are many very capable models in the market that you can connect to, but the problem is the infrastructure that you're managing them on," Abraham said.

Alongside model access, ExcelScale includes API key management, access controls, configurable model permissions, usage monitoring, and virtual private cloud deployment options. These controls are intended to help organizations manage access to AI services across multiple users and applications.

Abraham also discussed security requirements when applications connect to several AI services. He pointed to API key management, access restrictions, rate limits, monitoring, and controls over data flows as areas organizations need to manage, particularly in regulated environments.

Operational visibility is another part of the platform. ExcelScale's centralized dashboard tracks API requests, token consumption, and costs through a single interface.

"Every single call that they make, all of the tooling and the spend that never consolidates into one view, is a challenge for a business," Abraham said.

Abraham said organizations also need monitoring and audit trails when AI systems are deployed in production. A consolidated view gives administrators a way to track activity when different teams use multiple models and providers.

Cost control was another issue raised during the launch. "Six vendors, six invoices," Abraham said.

He said fragmented billing can make it harder to attribute AI spending to specific parts of a business and forecast costs. The platform combines cost and usage data with its access and model controls.

"As AI becomes part of everyday business operations, organizations need infrastructure that can evolve alongside the technology," Abraham said.