NVIDIA: Nemotron-3 Nano 30B A3B

nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B

Context

256K

Input

text

Output

text

Tool calling

Supported

About this model

NVIDIA Nemotron-3 Nano 30B A3B is NVIDIA's open reasoning model featuring a hybrid Mamba-Transformer Mixture-of-Experts (MoE) architecture. It consists of 23 Mamba-2 and MoE layers, plus 6 Attention layers. The model is built for building specialized AI agents, chatbots, and RAG systems, providing high compute efficiency. It natively supports configurable reasoning depth through a 'thinking budget' and acts as a general-purpose reasoning and chat model intended for English, coding languages, and several other supported languages.

Compare Text Models

Input modalities

text

Accepted as model input

Providers

Available routing options for this model through OneInfer.

Novita

Available

Input tokens

$0.050 / 1M tokens

Output tokens

$0.200 / 1M tokens

Routing

OneInfer optimized

Pricing

Current OneInfer pricing for this model.

Usage	Price
Input tokens	$0.050 / 1M tokens
Output tokens	$0.200 / 1M tokens

Performance

Published evaluation results associated with this model.

General Knowledge

MMLU-Pro78.3

Reasoning

GPQA (with tools)75

LiveCodeBench68.3

AIME25 (with tools)99.2

Agentic

SWE-Bench38.8

TauBench V2 (Average)49

BFCL v453.8

API example

Refer Documentation

curl https://api.oneinfer.ai/v1/ula/chat/completions \
  -H "Authorization: Bearer $JWT_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/NVIDIA-Nemotron-3-Nano-30B-A3B",
    "messages": [
      { "role": "user", "content": "Hello!" }
    ]
  }'