nvidiatext → text
NVIDIA: Nemotron 3 Nano 30B A3B
Canonical ID: nvidia/nemotron-3-nano-30b-a3b
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...
Artificial Analysis Benchmarks
VerifiedNo standardized Artificial Analysis benchmark scores recorded for this variant yet.
Token Pricing & Economics
Prompt Input (1M Tokens)
Input$0.06
Completion Output (1M Tokens)
Output$0.24
Prompt KV-Cache Read (1M Tokens)
Cached$0.01
Context Window Capacity
Context256k tokens
Supported Parameters & Features
frequency_penaltyinclude_reasoninglogit_biasmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p