Nemotron 3 Nano 30B

Nemotron 3 Nano 30B is NVIDIA's lightweight model in the Nemotron family, designed for efficient text processing tasks. As a 30 billion parameter model, it sits in the mid-range of model sizes while maintaining NVIDIA's focus on performance optimization. The model supports a 262,144 token context window and focuses exclusively on text processing without multimodal capabilities. Performance benchmarks show it generates 67.46 tokens per second with a time to first token of 303 milliseconds, indicating optimization for responsive applications. The model does not include tool calling functionality, keeping its design streamlined for core language tasks. Nemotron 3 Nano 30B targets use cases where balance between capability and efficiency matters, offering substantial context handling while maintaining faster response times compared to larger models in enterprise and production environments.