NVIDIA has launched four new NVIDIA NIM microservices to support sovereign AI efforts in Japan and Taiwan. These microservices are designed to help developers build and deploy high-performing GenAI applications that align with local values, laws, and interests.
The new microservices include Llama-3-Swallow-70B (trained on Japanese data), Llama-3-Taiwan-70B (trained on Mandarin data), and two RakutenAI 7B models for Chat and Instruct (trained on English and Japanese datasets). Compared to base LLMs, these models offer improved performance for regional language understanding, legal tasks, question-answering, and language translation and summarization.
The microservices will allow businesses, government agencies, and universities to host native LLMs in their environments, enabling the development of GenAI chatbots and assistants. The company also states that the microservices can provide up to 5x higher throughput, lowering the total cost of running the models in production and decreasing latency for better user experiences.
By using this site, you agree to allow SPEEDA Edge and our partners to use cookies for analytics and personalization. Visit our privacy policy for more information about our data collection practices.