Nemotron 3 Nano 30B A3B
by NVIDIAtext->text2 endpoints
Neura Intelligence Index
40.5/ 100
Rank
#240—
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
Prompt caching
Long context
Top-P sampling
Stop sequences
Deterministic seed
Modalities
Input
Text
Output
Text
Technical Specifications
Context Window
262.1K tokens
Max Output
235.9K tokens
Tokenizer
Other
Uptime (24h)
9970.6%
Parameters
32000000000B
License
nvidia_open_model_license_agreementOpen Weights
Country
🇺🇸US
Release Date
December 15, 202510mo
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biasmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_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 | 99.2 | verified |
reasoning
| Benchmark | Score | Source |
|---|---|---|
| GPQA Diamond | 75 | verified |
| Humanity's Last Exam | 15.5 | verified |
Performance Over Time
Pricing
Per 1M tokens
Input$0.06
Output$0.24
Cache read$0.01
Added
December 14, 2025
Last synced 11/7/2026