Fractal Analytics Limited (Fractal) has launched LLM Studio, an enterprise workbench that enables organisations to design, build, evaluate and operate domain-adapted language models using open-source models and NVIDIA AI infrastructure. The platform was announced on March 17, 2026, and will be shown at NVIDIA GTC 2026 from March 16 to 19 in San Jose. LLM Studio is intended for teams seeking greater control over governance, deployment and production management of customised language models.

LLM Studio includes AutoLLM, which helps create smaller specialised models by supporting model selection, synthetic data generation, customisation, evaluation and benchmarking, and LLMOps, which manages deployment, monitoring and governance. The workbench is designed to keep responses tied to an organisation's approved data and context to reduce hallucinations and improve reasoning quality. Resulting models remain proprietary to the organisation and can be used in agents and other generative AI applications.

The design leverages NVIDIA reference architectures, using NVIDIA NeMo for development workflow capabilities and NVIDIA NIM microservices for model hosting to standardise deployment across major cloud environments. Fractal plans to use NVIDIA Nemotron open models for development and positions LLM Studio as a way to deliver reliable production performance often at a fraction of the cost of larger foundation models. The chief executive officer, Pranay Agrawal, noted that enterprises are moving beyond experiments and require governed, cost predictable and reliable solutions, prompting the company to offer a platform that balances open-source flexibility with infrastructure consistency.

LLM Studio is aimed at enterprise users including those with limited or no coding experience and supports tailoring models to specific content and instruction following to strengthen reasoning for task-oriented applications. Fractal Analytics Limited is a publicly listed global enterprise AI company serving Fortune 500 organisations and employs over 5,000 professionals across key markets including the United States, the UK and India. The company highlighted recent recognition such as Top 100 India's Best Companies to Work For in 2025 and leadership placements in industry assessments, which supports its continued investment in enterprise AI capabilities.