Nova Micro

Nova Micro is built for speed, cost control, and predictable behavior in production. It keeps multi-turn conversations coherent, summarizes documents, answers questions with clear reasoning, and handles common coding and data tasks without heavy compute. The API supports function calls for retrieval and actions, streams tokens for responsive UX, and returns schema-true JSON so pipelines can parse outputs without brittle post-processing. It quantizes well for modest GPUs or edge devices, which makes fleet deployment practical. Teams typically use Nova Micro as the default engine for chat, support, and workflow automation, then route the hardest prompts to a larger Nova tier when extra depth is required.