NVIDIA model product
Nemotron 3 Nano is a 31.6B hybrid MoE model optimized for fast, long‑context agentic reasoning. It mixes Mamba‑2 and Transformer layers with a sparse MoE router (~3.6B active params per token) to deliver up to 4× higher throughput than Nemotron 2 and strong accuracy across math, coding, and tools. It supports a 1M‑token context window, offers Reasoning ON/OFF and a thinking‑budget to control costs, and ships with open weights, data, and RL tooling (NeMo Gym/RL). Released Dec 15, 2025 under the NVIDIA Open Model License, it’s built as the efficient backbone for multi‑agent systems at scale.
Updated Aug 10, 2026. Default version: Nemotron 3 Nano (30B A3B)
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version Nemotron 3 Nano (30B A3B).
| WMT24++ | 0.9 | 2 | 23 | 95.5% | C | |
| Arena-Hard v2 | 0.7 | 8 | 16 | 53.3% | C | |
| AIME 2025 | 1.0 | 10 | 114 | 92.0% | C | |
| SciCode | 0.3 | 18 | 18 | 0.0% | C | |
| Tau2 Airline | 0.5 | 20 | 23 | 13.6% | C | |
| Multi-Challenge | 0.4 | 24 | 29 | 17.9% | C | |
| Terminal-Bench | 0.1 | 24 | 25 | 4.2% | C | |
| MMLU-ProX | 0.6 | 25 | 32 | 22.6% | C | |
| Tau2 Retail | 0.6 | 26 | 26 | 0.0% | C | |
| Tau2 Telecom | 0.4 | 33 | 35 | 5.9% | C | |
| LiveCodeBench v6 | 0.7 | 34 | 53 | 36.5% | C | |
| MMLU-Pro | 0.8 | 58 | 129 | 55.5% | C | |
| Humanity's Last Exam | 0.2 | 69 | 92 | 25.3% | C | |
| SWE-Bench Verified | 0.4 | 96 | 104 | 7.8% | C | |
| GPQA | 0.8 | 107 | 233 | 54.3% | C |
Preference and agent-evaluation signals from published Arena datasets.
The default version has no published Arena rows, or its source alias has not been resolved.
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| Kenari | nemotron-3-nano-30b-a3b | global | N/A | N/A | 262.1K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Nemotron 3 Nano (30B A3B) | 45.1 | 32B | 262.1K | 262.1K | Yes | NVIDIA Open Model License Agreement |
A concise description based on the published model registry.
Nemotron 3 Nano is a 31.6B hybrid MoE model optimized for fast, long‑context agentic reasoning. It mixes Mamba‑2 and Transformer layers with a sparse MoE router (~3.6B active params per token) to deliver up to 4× higher throughput than Nemotron 2 and strong accuracy across math, coding, and tools. It supports a 1M‑token context window, offers Reasoning ON/OFF and a thinking‑budget to control costs, and ships with open weights, data, and RL tooling (NeMo Gym/RL). Released Dec 15, 2025 under the NVIDIA Open Model License, it’s built as the efficient backbone for multi‑agent systems at scale.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Nemotron 3 Nano.
Nemotron 3 Nano's default version was released on Dec 15, 2025.
No official standard PAYG price is currently available for Nemotron 3 Nano.
Nemotron 3 Nano is published under NVIDIA in the model registry.
The default version has a 262.1K token context window.
Yes. The default version is marked as open weight under NVIDIA Open Model License Agreement .
1 published provider offerings are linked to the default version.
Nearby ranked alternatives include Sarvam 30B, Qwen2.5 Coder 32B, Gemma 3 27B.