← Back to Directory

NVIDIA: Nemotron 3 Nano 30B A3B

NVIDIA Developer Architecture Profile

OpenRouter ↗HuggingFace ↗
Intelligence (ELO)1150Chatbot Arena Verified
Max Context262,144Tokens
API Cost / 1M$0.25Blended Prompt + Completion

Model Capabilities

    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 open with open-weights, datasets and recipes so developers can easily customize, optimize, and deploy the model on their infrastructure for maximum privacy and security.

    Granular Pricing Matrix

    Input Tokens (Prompt)$0.05 / 1M
    Output Tokens (Completion)$0.20 / 1M

    Pricing data via OpenRouter. Sync: 3/16/2026

    Evaluate Competitors

    VS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs ByteDance Seed: Seed-2.0-LiteVS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs Qwen: Qwen3.5-35B-A3BVS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs Qwen: Qwen3.5-FlashVS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs MiniMax: MiniMax M2.5 (free)VS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs MiniMax: MiniMax M2.5VS Engine MatchupNVIDIA: Nemotron 3 Nano 30B A3B vs StepFun: Step 3.5 Flash (free)