Malaysia debuts first multimodal large language model

Malaysia debuts first multimodal large language model

Tech in Asia·2025-08-12 20:00

Prime Minister Datuk Seri Anwar Ibrahim launched ILMU, described as Malaysia’s first multimodal large language model (LLM), at the ASEAN AI Malaysia Summit 2025.

Developed by YTL AI Labs in collaboration with Universiti Malaya, ILMU can handle text, voice, and images, supporting applications across Malaysian industries.

YTL AI Labs, part of YTL Power International Berhad, said ILMU can process Malay, Manglish, and local dialects, and is hosted on local cloud infrastructure for data residency.

The company claims ILMU matches leading global models like GPT-4 and Llama in several international benchmarks, and achieves the highest score among all frontier models in Bahasa Melayu language understanding.

YTL AI Labs and MDEC also announced the ILMU AI Accelerator Programme, offering Malaysian startups, SMEs, and global solution providers benefits including RM5 million in free ILMU API credits, MD Status, and talent facilitation.

Partners at launch included Aco Tech, Ryt Bank, Astro, Media Prima, Yes, Carsome, Swipey, Vision Machina, Vistel, TeeniAI, TrustAI, and Mesolitica.

ILMUchat, a consumer-focused AI chatbot, will be available for early access on September 16, 2025.

.source-ref{font-size:0.85em;color:#666;display:block;margin-top:1em;}a.ask-tia-citation-link:hover{color:#11628d !important;background:#e9f6f5 !important;border-color:#11628d !important;text-decoration:none !important;}@media only screen and (min-width:768px){a.ask-tia-citation-link{font-size:11px !important;}}

🔗 Source: YTL AI Labs

🧠 Food for thought

1️⃣ Local language AI models emerge as strategic assets in Southeast Asian competition

ILMU’s launch positions Malaysia in an intensifying regional race to become Southeast Asia’s leading AI hub, where countries are leveraging their unique cultural and linguistic advantages.

Singapore currently dominates the region with $741 million committed over five years to AI development, while other ASEAN nations are developing competing strategies to attract investment and establish technological sovereignty1.

Southeast Asian developers have been specifically creating localized large language models to address the underrepresentation of regional languages in predominantly English-based AI tools2. ILMU’s fluency in Bahasa Melayu, Manglish, and local dialects like Kelate directly addresses this gap, with the model achieving the highest score among frontier models in Bahasa Melayu language understanding.

This localization strategy reflects a broader trend where over 1,200 languages spoken across Southeast Asia represent both a challenge and opportunity for AI developers seeking to serve diverse populations2. By claiming world-leading performance in Bahasa Melayu, YTL AI Labs is positioning Malaysia to capture value from linguistic differentiation rather than competing purely on technical specifications against global models.

2️⃣ Malaysia’s coordinated AI infrastructure investments show long-term strategic planning

ILMU’s development represents the culmination of Malaysia’s systematic approach to building AI capabilities through sustained government investment and strategic partnerships.

The Malaysian government has been investing RM10 billion annually in skills training and workforce development, with the establishment of the National Artificial Intelligence Office (NAIO) to drive the country’s AI agenda34. This infrastructure spending created the foundation for YTL AI Labs to develop ILMU with Universiti Malaya collaboration.

Malaysia has also secured strategic partnerships with major technology companies, including Microsoft’s cloud region launch and Petronas-Microsoft collaboration to establish the country as a regional AI hub5. YTL’s hosting of ILMU on their own YTL AI Cloud infrastructure demonstrates how these investments in local capabilities reduce dependence on foreign cloud providers.

The ILMU AI Accelerator Programme offering RM5 million in free API credits to Malaysian startups shows how the government is using the homegrown model to stimulate local AI adoption, following a pattern where 70% of Malaysian business leaders believe AI is crucial for competitiveness6.

……

Read full article on Tech in Asia

SE Asia Technology