NVIDIA has announced the deployment of its first on-device small language model (SLM) for game characters, promising a more dynamic and immersive gameplay experience. The model, Nemotron-4 4B Instruct, is optimized for faster deployment and better performance in specific tasks. It will be available for both cloud and on-device deployment, with the first game to showcase its capabilities being ‘Mecha BREAK’ by Amazing Seasun Games.
