Nemotron 3 Nano 30B A3B
by NVIDIAtext->text3 endpoints
Neura Intelligence Index
40.0/ 100
Rank
#193—
Confidence
mediumBenchmarks
6(15% coverage)
Overview
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...
Capabilities
Text generation
Tool use / Function calling
Structured output
Extended reasoning
Long context
Top-P sampling
Modalities
Input
Text
Output
Text
Technical Specifications
Context Window
262.1K tokens
Max Output
228K tokens
Tokenizer
Other
Uptime (24h)
9965.4%
Parameters
32000000000B
License
nvidia_open_model_license_agreementOpen Weights
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Benchmark Performance
coding
| Benchmark | Score | Source |
|---|---|---|
| SWE-bench Verified | 38.8 | verified |
| SciCode | 33.3 | verified |
| TerminalBench | 8.5 | verified |
math
| Benchmark | Score | Source |
|---|---|---|
| AIME 2025 |
Performance Over Time
Pricing
Per 1M tokens
Input$0.05
Output$0.20