Skip to content
Yumeforge

NVIDIA

Nemotron 3 Nano Omni Vision

No ratings yet
  • Studio: Describe
  • Makes: Text
  • Task: Image to text

Get captions, prompts and answers from an image with Nemotron 3 Nano Omni Vision. Pay per result with credits. No subscription.

Open in studio

from 3 credits

Cafe table with a croissant, cappuccino and notebook by the window
Example from the Describe studio, made with Z-Image Turbo

What you can set

  • PromptRequired

    Text

    Text prompt to send to the model. English only.

  • ImageRequired

    Upload an image

    Image URL to analyze with the model.

  • System prompt

    Text

    Optional system prompt to steer the model.

  • Max tokens

    Number

    Maximum number of tokens to generate.

  • Reasoning mode

    Choice

    Whether the model should emit an explicit reasoning trace.

    • no_think (default)
    • think

  • Temperature

    Number

    Sampling temperature. Lower values are more deterministic.

  • Top p

    Number from 0 to 1

    Nucleus sampling probability mass.

Reviews

From creators who made something with Nemotron 3 Nano Omni Vision.

No ratings yet. Ratings come only from people who made something with this model.