← All models View on HuggingFace →
N
NVIDIA: Nemotron 3 Nano 30B A3B
nvidia/nemotron-3-nano-30b-a3b
Tool useReasoningStreaming
NVIDIA Nemotron 3 Nano 30B A3B is a small language MoE model with highest compute efficiency and accuracy for developers to build specialized agentic AI systems. The model is fully...
Pricing
Input
$0.05 / 1M
Output
$0.2 / 1M
Specs
Context
256,000 tokens
Input
text
Output
text
Released: 2025-12
Supported parameters
include_reasoningmax_tokensreasoningseedtemperaturetool_choicetoolstop_p
Open weights · HuggingFace
1,695,640 downloads/mo
744 likes
other text-generation