N
Nemotron 3 Nano 30B A3BOPEN
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.
Prices update automatically — checked daily against provider list prices.
- $0.11 per 1M tokens, assuming 3 input for 1 output the convention used by Artificial Analysis
- cheaper than 77% of the 200 open-weight models tracked here
- 8 of 12 NVIDIA models tracked here, from cheapest
Benchmarks & performance
Independent benchmark scores for Nemotron 3 Nano 30B A3B, measured by Artificial Analysis. Higher is better. Measurement mode: Reasoning.
About this model
Nemotron 3 Nano 30B A3B is an open-weight AI model by NVIDIA. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.
Frequently asked questions
What is Nemotron 3 Nano 30B A3B?
Nemotron 3 Nano 30B A3B is an AI language model from NVIDIA. It is open-weight: you can download it and run it on your own hardware, for free.
Is Nemotron 3 Nano 30B A3B free?
The weights are free and open — you can self-host Nemotron 3 Nano 30B A3B and pay nothing per token. If you prefer a hosted API, list prices are $0.06 per million input tokens and $0.24 per million output tokens.
What is Nemotron 3 Nano 30B A3B good at?
Independent benchmarks from Artificial Analysis give it GPQA 75.7%, MMLU-Pro 79.4%, Humanity's Last Exam 11.4%, Long Context Reasoning 38%, LiveCodeBench 74.1%, SciCode 30.6%, AIME 2025 91%, IFBench 71.1%, τ²-Bench 40.9%, τ-Bench Banking 6%, Terminal-Bench 6.7%, Terminal-Bench Hard 13.6%. It is particularly used for code generation.
How fast is Nemotron 3 Nano 30B A3B?
It generates about 238.9 tokens per second, with a median 0.4s delay before the first token. Measured independently by Artificial Analysis.
Can I self-host Nemotron 3 Nano 30B A3B?
Yes. Nemotron 3 Nano 30B A3B has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.