Blog

Puter.js now supports Nvidia Nemotron, giving developers free, unlimited access to Nvidia's efficient language models.

What is Nemotron?

Nemotron is NVIDIA's foundation model family specifically designed for building efficient, accurate, and specialized agentic AI systems. These models excel at advanced reasoning, coding, visual understanding, and safety-focused tasks. What sets Nemotron apart is its open-source approach with transparent training data and broad platform support, making it easier to create and deploy trustworthy, high-performance AI agents. Through pruning of larger models and optimization with NVIDIA TensorRT-LLM, Nemotron delivers superior compute efficiency while achieving high accuracy on leading benchmarks—perfect for both edge devices and cloud deployment.

Available Models

Nemotron brings three powerful options optimized for different use cases:

• Nemotron Nano 9B: Efficient text generation for everyday tasks
• Nemotron Nano 12B Vision: Multimodal capabilities including image analysis
• Nemotron Super: Advanced reasoning and complex problem-solving

All Nemotron models are available through Puter.js with no API keys, no usage limits, and no infrastructure setup required.

Quick Examples

Text generation:

puter.ai.chat("Explain neural networks simply", { 
    model: 'nvidia/nemotron-nano-9b-v2' 
});

Image analysis:

puter.ai.chat(
    "Describe this image",
    "https://assets.puter.site/doge.jpeg",
    { model: 'nvidia/nemotron-nano-12b-v2-vl' }
);

Complex reasoning:

puter.ai.chat(
    "Analyze quantum computing's impact on encryption",
    { model: 'nvidia/llama-3.3-nemotron-super-49b-v1.5' }
);

Get Started

Add one script tag to your HTML:

<script src="https://js.puter.com/v2/"></script>

No sign-ups, no API keys, no server setup. Start building with Nvidia Nemotron immediately.

Learn more:

• Nvidia Nemotron Tutorial
• Puter.js Documentation