Skip to main content

Image & Audio Components

Generate and analyze images, play or save audio and images directly on the canvas, and capture screenshots of the Grasshopper canvas or Rhino viewports (SmartHopper → G. Media).

Media components

One-step image generation & vision (G. Media)​

IconComponentNicknameWhat it doesPreview
AI Text To Image iconAI Text To ImageAIText2ImgGenerates an image from a text Prompt using an image-generation model, with options for Size and Aspect Ratio. Outputs the generated Image and any provider-returned Revised Prompt.ai-text-to-image
AI Image To Text iconAI Image To TextAIImg2TextDescribes or answers questions about an Image (Bitmap, file path, URL, or base64) using a vision-capable model, guided by an optional Prompt.ai-image-to-text
Input → Output media components

When you want to combine media with other inputs in a single call, use Image to AI (Img2AI) and Audio to AI (Audio2AI) in B. Input together with AI to Image (AI→Img) and AI to Speech (AI→Speech) in C. Output.

On-canvas viewers (G. Media)​

IconComponentNicknameWhat it doesPreview
Image Viewer iconImage ViewerImgViewDisplays an image directly on the Grasshopper canvas and saves it to disk (.png, .jpg, .bmp, .gif, .tiff) when Save? is triggered.image-viewer
Audio Viewer iconAudio ViewerAudioViewPreviews and plays audio directly on the canvas (built-in Play button) and saves it to disk (.mp3, .wav, .m4a, .aac, .flac, .ogg, .wma, .opus) when Save? is triggered.audio-viewer

Canvas & viewport capture (G. Media)​

IconComponentNicknameWhat it doesPreview
Canvas To Image iconCanvas To ImageCanvas2ImgCaptures a screenshot of the visible Grasshopper canvas when Run? is set to True (bounded by Max Width and Max Height, default 1920×1080) and outputs the Image along with its pixel Width and Height.canvas-to-image
Viewport To Image iconViewport To ImageViewport2ImgCaptures a screenshot of the active or named Rhino viewport (View Name) when Run? is set to True (bounded by Width and Height, default 1024×1024) and outputs the Image, Width, Height, and captured View Name.viewport-to-image

Wire Viewport To Image into AI Image To Text (or Image to AI → AI to Text) to ask a vision model about your 3D geometry — for example, checking massing proportions or diagnosing a broken mesh preview.

Modality fallback

If your selected model does not natively accept image or audio inputs, SmartHopper's Modality fallback setting (SmartHopper → Settings → General) can automatically convert the media through a capable model first — see Settings reference.