Birla AI Labs, an artificial intelligence initiative and venture studio under the Aditya Birla Group, has launched Yug-1, its first time series foundation model designed to analyse historical data, identify patterns and generate forecasts across business and industrial applications.
The India-developed model focuses on time-dependent information, including demand and supply fluctuations, financial indicators, equipment performance and infrastructure usage. Unlike large language models that primarily process text, time series foundation models are trained to recognise patterns in data collected over successive periods.
Yug-1 is designed to analyse interconnected datasets, forecast multiple variables and identify anomalies that may require attention. Such capabilities can support applications ranging from inventory planning and financial forecasting to industrial monitoring and energy management.
Ananya Birla, Founder and Chairperson of Birla AI Labs, announced the development, highlighting the importance of applying artificial intelligence to data generated through everyday business and economic activity.
“Most of the conversation around AI today is about language. But so much of the data that actually runs businesses and economies moves through time,” Birla wrote in a LinkedIn post.
According to the company, Yug-1 performed better than other first-generation time series foundation models on the GiftEval benchmarking framework and delivered competitive results against newer models. These performance claims were shared by Birla AI Labs and have not been independently verified.
The company said its research team completed the development process, from training to deployment, in less than six months. Yug-1 has been made available through GitHub and Hugging Face, allowing developers and researchers to evaluate the model and explore potential applications.
Birla AI Labs has described Yug-1 as Asia's first time series foundation model developed for real-world applications. The designation remains a company claim.
The launch expands the laboratory's research focus beyond conversational AI towards predictive systems capable of supporting decisions involving operational and financial data.
For businesses, time series forecasting can help estimate future demand, monitor changing consumption patterns and identify unusual activity in equipment or infrastructure. In retail and marketing operations, similar analytical approaches can support sales forecasting, inventory decisions and resource planning.
The Birla AI Labs research team has previously presented its work at NeurIPS 2025, ICML 2026 and the World Summit AI, alongside engagements with universities in India and overseas.
“When we started the research arm of Birla AI Labs, the belief was simple: research only matters when it reaches the real world,” Birla said.
The company plans to present Yug-1 at NeurIPS 2026 and engage with international research institutions. Birla has also invited developers to test the model and share feedback to inform its future development.
The release represents Birla AI Labs' entry into foundation models for time-dependent data, with broader adoption and performance across commercial deployments yet to be established.
Disclaimer: This article may include information derived from interviews, press releases, public statements, research, company communications and other publicly available or third-party sources. Such material may be summarised, paraphrased or contextualised for journalistic and editorial purposes. All rights in third-party content remain with their respective owners.