Japanese companies are using open models and datasets for language, telecom, energy, manufacturing and enterprise workflows.

NVIDIA announced that Japanese enterprises, startups and research institutions are using NVIDIA Nemotron open models, data and libraries to build AI models and applications for Japan-specific language, industry and workforce needs.

Open models give organizations a way to customize, deploy and govern AI systems within their own environments. In Japan, these tools are being used as organizations address workforce changes, productivity requirements and demand for AI systems adapted to local industries.

Building specialized AI for Japan with NVIDIA Nemotron

Developers in Japan are using NVIDIA Nemotron open models and datasets to build AI systems for industry and public-sector applications.

Institute of Science Tokyo developed its Swallow family of open foundation models using NVIDIA Nemotron datasets and the NVIDIA NeMo software stack for continual pretraining and post-training. Swallow models are designed to improve Japanese-language and reasoning performance while retaining English, math and coding capabilities. Enterprises are adapting Swallow for uses such as financial-document translation and asset-management report generation.

SB Intuitions Corp., SoftBank Corp.’s generative AI research subsidiary, trained its Sarashina series of generative AI models using NVIDIA Nemotron, including the NVIDIA NeMo RL and Megatron-LM libraries. Sarashina3 mini has been selected by Japan’s Digital Agency for specialized AI use cases. SoftBank Corp. has also developed and deployed a Large Telecom Model using Sarashina and NVIDIA Nemotron for autonomous telecom network operations.

Stockmark released a Japanese-language document-understanding model based on the NVIDIA Nemotron 3 Nano Omni model. The company is also developing enterprise knowledge applications using NVIDIA NeMo Retriever and the Nemotron-Personas-Japan dataset for customers in Japan’s manufacturing, energy and chemical industries through Japan’s Generative AI Accelerator Challenge national project.

Applying NVIDIA Nemotron in Japan’s industries

Japanese companies are using NVIDIA Nemotron to support services, productivity and workforce-related applications.

AI and robotics startup avatarin is using NVIDIA Nemotron open models and NVIDIA NeMo to develop Japanese-language speech and reasoning capabilities for enterprise AI agents. NVIDIA HGX B300 systems provide private AI infrastructure for analyzing customer conversations and accessing enterprise knowledge, while NVIDIA Jetson supports edge AI functions, including digital avatar systems being deployed at airports and other locations in Japan.

ENEOS Holdings is using NVIDIA Nemotron open models with the NVIDIA AI-Q Blueprint and NVIDIA ALCHEMI NIM microservices for agentic AI workflows in energy and materials R&D. Researchers are using these technologies to combine technical document search, vision and language understanding and simulation-backed molecular screening for materials exploration, including immersion-cooling fluids and catalysts.

NTT DATA, an operating subsidiary of NTT, used NVIDIA Nemotron-Personas-Japan to augment training data for its proprietary tsuzumi 2 model, improving question-answering accuracy and responses to questions that require additional knowledge. The company is also evaluating a scalable multi-agent framework using NVIDIA Agent Toolkit, including NVIDIA Nemotron, to route tasks to appropriate models for enterprise workflows.

Hitachi is developing physical AI technologies using NVIDIA Nemotron and NVIDIA Cosmos open models, along with its own IT and operational technology domain knowledge. In a multi-agent orchestration platform, these technologies are intended to connect and coordinate IT and OT operations across enterprise workflows.

Sakana AI is working with NVIDIA to integrate NVIDIA Nemotron into its Fugu model-orchestration platform. Fugu is designed to select models for specific tasks in agentic AI workflows, helping developers manage accuracy, performance and cost across open and proprietary AI models.

Open and customizable deployment

Nemotron models are released with open weights, datasets and recipes, allowing organizations to inspect, customize and deploy models for domain-specific workflows.

Developers can use NVIDIA NeMo to customize, evaluate and optimize models for specific applications and deploy them in environments aligned with regulatory, sovereignty and data-localization requirements.

Nemotron models are available on Hugging Face, ModelScope, OpenRouter and build.nvidia.com as NVIDIA NIM microservices, as well as through NVIDIA Cloud Partners, inference platforms and cloud service providers.

For more information, visit nvidia.com.