NVIDIA Alpamayo

Nemotron 3 Nano is a compact, open large language model in NVIDIA’s Nemotron 3 family, designed for efficient agentic reasoning, conversational AI, and coding tasks. It uses a hybrid Mixture-of-Experts Mamba-Transformer architecture that activates only a small subset of parameters per token, enabling low-latency inference while maintaining strong accuracy and reasoning performance. It has approximately 31.6 billion total parameters with around 3.2 billion active (3.6 billion including embeddings), allowing it to achieve higher accuracy than previous Nemotron 2 Nano while using less computation per forward pass. Nemotron 3 Nano supports long-context processing of up to one million tokens, enabling it to handle large documents, multi-step workflows, and extended reasoning chains in a single pass. It is designed for high-throughput, real-time execution, excelling in multi-turn conversations, tool calling, and agent-based workflows where tasks require planning, reasoning, and more.

Software is what turns a vehicle into an intelligent machine. The NVIDIA DRIVE™ Software stack is open, empowering developers to efficiently build and deploy a variety of state-of-the-art AV applications, including perception, localization and mapping, planning and control, driver monitoring, and natural language processing. The foundation of the DRIVE Software stack, DRIVE OS is the first safe operating system for accelerated computing. It includes NvMedia for sensor input processing, NVIDIA CUDA® libraries for efficient parallel computing implementations, NVIDIA TensorRT™ for real-time AI inference, and other developer tools and modules to access hardware engines. The NVIDIA DriveWorks® SDK provides middleware functions on top of DRIVE OS that are fundamental to autonomous vehicle development. These consist of the sensor abstraction layer (SAL) and sensor plugins, data recorder, vehicle I/O support, and a deep neural network (DNN) framework.