AI Tools & Products News

Malaysia Unveils ILMU: Its First Homegrown Multimodal AI Model

Malaysia Prime Minister Datuk Seri Anwar Ibrahim officially launched ILMUIntelek Luhur Malaysia Untukmu—the nation’s first entirely homegrown multimodal large language model (LLM).

A Model Tailored for Malaysians

Developed by YTL AI Labs (a subsidiary of YTL Power International Bhd), in partnership with the University of Malaya, ILMU is built, owned, and operated entirely within Malaysia, ensuring the nation retains strategic AI sovereignty. 

Designed to process text, voice, and vision, ILMU fluently handles Bahasa Melayu, Manglish, and Malaysian dialects such as Kelantanese, leveraging local linguistic and cultural understanding.

Benchmark Performance

LMU achieved top scores on the Malay MMLU benchmark, outperforming DeepSeek V3, GPT-4o, and GPT-5 in Bahasa Melayu understanding, while matching global frontier models like GPT-4o and Llama 3.1 in overall

The model embodies its name’s core principles: Intelek (intelligence), Luhur (integrity/ethics), Malaysia (cultural authenticity), and Untukmu (designed for all Malaysians).

Availability & pricing

ILMUchat, a consumer-facing chatbot powered by ILMU, is set for early access on Malaysia Day, September 16, 2025. Public registration is open now via the official site.

For businesses, ILMU API access is already available.

YTL AI Labs, in partnership with MDEC, has launched the ILMU AI Accelerator Programme, offering up to USD 1 million (or RM5 million in free API credits, per some sources) to startups and innovators, plus support for Malaysia Digital status and talent passes.

Over 10 industry partners—including Aco Tech, Carsome, Astro, Ryt Bank, Media Prima, Yes 5G, Swipey, Vision.Machina, Vistel, TeeniAI, TrustAI, and Mesolitica are already onboarded to explore ILMU’s capabilities across sectors.

ILMU stands out as Malaysia’s first fully homegrown multimodal AI model, built to rival global leaders like OpenAI’s GPT-4o, Meta’s Llama 3.1, and DeepSeek V3.

While those international models excel in broad, multilingual capabilities, ILMU’s competitive edge lies in its deep cultural and linguistic understanding of Malaysia—including fluency in Bahasa Melayu, Manglish, and regional dialects.

Unlike its foreign counterparts, ILMU is developed, hosted, and owned entirely within Malaysia, giving it advantages in data sovereignty, local compliance, and cultural relevance.

News Gist:

Malaysia has launched ILMU, its first homegrown multimodal AI model, rivaling GPT-4o and Llama 3.1.

Fluent in local languages and dialects, ILMU offers free early access via ILMUchat on September 16, with API access for businesses already open.

Leave a Reply

Your email address will not be published. Required fields are marked *

AI Binger
Privacy Overview

This website uses cookies so that we can provide you with the best user experience possible. Cookie information is stored in your browser and performs functions such as recognising you when you return to our website and helping our team to understand which sections of the website you find most interesting and useful.