KUALA LUMPUR, March 14 (Bernama) -- H2O.ai, a leader in open-source generative artificial intelligence (GenAI) and predictive AI platforms, has launched H2O Enterprise LLM Studio, running on Dell infrastructure.
According to a statement, this new offering provides Fine-Tuning-as-a-Service for businesses to securely train, test, evaluate, and deploy domain-specific AI models at scale using their data.
H2O.ai Chief Executive Officer and Founder, Sri Ambati said distilling and fine-tuning AI models are transforming enterprise workflows, making operations smarter and more efficient.
“H2O Enterprise LLM Studio makes it simple for businesses to build domain-specific models without complexity,” he said.
Developed by top Kaggle Grandmasters, the Enterprise LLM Studio automates the entire large language model (LLM) lifecycle, from data curation to deployment, and supports various open-source and multimodal LLMs such as DeepSeek, Llama, and H2O Danube. By leveraging model distillation and fine-tuning techniques, H2O.ai customers can achieve cost reductions and improved inference speeds.
Key features include model distillation, no-code fine-tuning, advanced optimisation, and scalable AI deployment, all designed to improve efficiency and scalability.
With fine-tuned open-source LLMs, businesses can cut expenses by up to 70 per cent and reduce inference time by 75 per cent. H2O.ai’s solution offers privacy, flexibility and avoids vendor lock-in, making it an ideal tool for enterprises looking to scale AI securely and cost-effectively.
H2O.ai will showcase its AI innovations at NVIDIA GTC, including sessions on model distillation and AI applications in the telecom sector. Industry leaders such as AT&T Chief Data and AI Officer, Andy Markus and Vice President of Data Science, AI & Automation, Mark Austin will participate in discussions on AI-driven transformation.
-- BERNAMA
