NVIDIA: Nemotron Nano 9B V2 (free)

NVIDIA-Nemotron-Nano-9B-v2 is a large language model (LLM) trained from scratch by NVIDIA, specifically engineered as a unified solution for both complex reasoning and general non-reasoning tasks. This model excels at responding to user queries by first generating a detailed reasoning trace, then concluding with a precise final response. Its unique architecture allows for flexible control over its reasoning capabilities via system prompts, enabling users to tailor its output to their specific needs. This free-to-use model boasts a substantial 128K token context window and a maximum output of 4K tokens, making it suitable for extensive conversations and detailed responses. It supports advanced capabilities such as functions, code generation, and streaming, enhancing its utility for developers and power users. With a pricing of $0.00 per 1M tokens for both input and output, NVIDIA Nemotron Nano 9B V2 offers an exceptional, cost-free AI experience. Discover its full potential on Multi AI.