NVIDIA model product
Nemotron 3 Nano
6B hybrid MoE model optimized for fast, long‑context agentic reasoning.
Updated Aug 12, 2026. Default version: Nemotron 3 Nano (30B A3B)
LLMBoard score35.2Nemotron 3 Nano (30B A3B)
Coverage100%15 benchmark families
Context window262.1KTokens
Official input priceN/AOfficial price unavailable
Nemotron 3 Nano Specifications
Technical details for the model's default version.
- Version
- Nemotron 3 Nano (30B A3B)
- Released
- Dec 15, 2025
- Knowledge cutoff
- Nov 28, 2025
- Parameters
- 32B
- Context window
- 262.1K
- Max output
- 262.1K
- Inputs
- text
- Outputs
- text
- Open weights
- Yes
- License
- NVIDIA Open Model License Agreement
Nemotron 3 Nano Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Nemotron 3 Nano Benchmark Results
Benchmark scores for Nemotron 3 Nano (30B A3B).
| WMT24++ | 86.2% | 02 | 23 | 95.5% | C | |
| Arena-Hard v2 | 67.7% | 08 | 16 | 53.3% | C | |
| AIME 2025 | 99.2% | 10 | 114 | 92.0% | C | |
| SciCode | 33.3% | 19 | 19 | 0.0% | C | |
| Tau2 Airline | 48.0% | 20 | 23 | 13.6% | C | |
| Multi-Challenge | 38.5% | 24 | 29 | 17.9% | C | |
| Terminal-Bench | 8.5% | 24 | 25 | 4.2% | C | |
| MMLU-ProX | 59.5% | 25 | 32 | 22.6% | C | |
| Tau2 Retail | 56.9% | 26 | 26 | 0.0% | C | |
| Tau2 Telecom | 42.2% | 33 | 35 | 5.9% | C | |
| LiveCodeBench v6 | 68.3% | 34 | 53 | 36.5% | C | |
| MMLU-Pro | 78.3% | 58 | 129 | 55.5% | C | |
| Humanity's Last Exam | 15.5% | 70 | 93 | 25.0% | C | |
| SWE-Bench Verified | 38.8% | 97 | 105 | 7.7% | C | |
| GPQA | 75.0% | 108 | 234 | 54.1% | C |
Nemotron 3 Nano Arena Results
Preference and agent-evaluation results for the default version.
| text factuality | overall | 149 | 1363.8 | 15,298 | N/A | |
| text | overall | 181 | 1348.8 | 15,587 | N/A | |
| text style control | overall | 240 | 1315.1 | 15,587 | N/A |
Nemotron 3 Nano Runtime Performance
Provider-specific output speed and catalog latency for Nemotron 3 Nano (30B A3B). Runtime does not affect the capability score.
| DeepInfra | 275.936 tok/s | 8.963 s | 262.1K | 262.1K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Nemotron 3 Nano Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- N/A
- Official provider
- N/A
- Lowest third-party
- From $0.05 input, $0.20 output per 1M via Pioneer
- Tracked offerings
- 10
| Nvidia | nvidia/nemotron-3-nano-30b-a3b | global | N/A | N/A | 131.1K | |
| Kenari | nemotron-3-nano-30b-a3b | global | N/A | N/A | 262.1K | |
| Pioneer | nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B-BF16 | global | $0.05 | $0.20 | 262.1K | |
| Kilo Gateway | nvidia/nemotron-3-nano-30b-a3b | global | $0.05 | $0.20 | 262.1K | |
| Vercel AI Gateway | nvidia/nemotron-3-nano-30b-a3b | global | $0.05 | $0.24 | 262.1K | |
| OpenRouter | nvidia/nemotron-3-nano-30b-a3b | global | $0.05 | $0.20 | 262.1K | |
| Deep Infra | nvidia/Nemotron-3-Nano-30B-A3B | global | $0.05 | $0.20 | 262.1K | |
| Nebius Token Factory | nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B | global | $0.06 | $0.24 | 32K | |
| NanoGPT | nvidia/nemotron-3-nano-30b-a3b | global | $0.17 | $0.68 | 256K |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Nemotron 3 Nano Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Nemotron 3 Nano (30B A3B) | 35.2 | 32B | 262.1K | 262.1K | Yes | NVIDIA Open Model License Agreement |
What is Nemotron 3 Nano?
Key information about Nemotron 3 Nano and its available data.
6B hybrid MoE model optimized for fast, long‑context agentic reasoning. 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.
Data as of 2026-08-11.
Nemotron 3 Nano vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Nemotron 3 Nano
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
NV
Nemotron Nano 9B
NVIDIA
NV
Llama 3.1 Nemotron Ultra 253B
NVIDIA
NV
Nemotron 3 Super
NVIDIA
NV
Llama 3.3 Nemotron Super 49B
NVIDIA
NV
Llama 3.1 Nemotron Nano 8B
NVIDIA
NV
Nemotron 3 Ultra
NVIDIA
FAQ
Common questions about Nemotron 3 Nano.
When was Nemotron 3 Nano released?
Nemotron 3 Nano's default version was released on Dec 15, 2025.
How much does Nemotron 3 Nano cost?
No official standard PAYG price is currently available for Nemotron 3 Nano. The lowest tracked third-party offer starts at $0.05 input and $0.20 output via Pioneer.
Who created Nemotron 3 Nano?
Nemotron 3 Nano was created by NVIDIA.
What is the context window for Nemotron 3 Nano?
The default version has a 262.1K token context window.
Is Nemotron 3 Nano open weight?
Yes. The default version is marked as open weight under NVIDIA Open Model License Agreement .
How many API providers offer Nemotron 3 Nano?
10 provider offerings are linked to the default version.
What models should I compare Nemotron 3 Nano with?
Nearby ranked alternatives include Qwen3 Coder 480B A35B, MiniMax M1 40K, Qwen3 VL 32B.