Visual replies in chat
Learn how generated media fits alongside maps, charts, QR codes, citations, and other visual chat replies.
Turn on a media tool
You can reach the media tools in three ways:
Tool mode Off turns both media tools off.
/create-image, /create-video, and other explicitly selected tools are not available with Auto in the model picker, Claude Opus 5.5, or Claude Sonnet 5.5. With those, turn on Manual > Image generation or Manual > Video generation, or use Auto tool mode and ask for the image or video. You can also choose another named model for that request.Create an image
1
Turn on Image generation
Use Auto tool mode, choose Manual > Image generation, or type
/create-image.2
Describe the image
Include subject, style, composition, lighting, and whether you want a square, portrait, or landscape image.
3
Review the result
Save, download, or keep refining the image in the chat.
Edit an image
The Image generation tool can also change an existing image. It edits one source image at a time.- Attach the image you want to change, or refer to an image Babbily generated earlier in the same chat.
- Describe the change, for example
Make the background a soft blue gradient and remove the coffee cup. - If the chat has several images, say which one you mean. Babbily may ask you to confirm.
Create a video
1
Turn on Video generation
Use Auto tool mode, choose Manual > Video generation, or type
/create-video.2
Set duration and shape
When you select the tool, the composer shows Video duration (3 to 10 seconds) and Video aspect ratio (landscape, portrait, or square).
3
Describe the motion
Video generation works best with clear motion, subject, and pacing instructions.
- Animate an image. Attach one image and describe how it should move.
- Edit a video. Attach one video and describe the change. Duration and aspect ratio settings do not apply when you edit a video.
If your previous media model is missing
On October 6, 2026, image and video models stopped being separate choices in the model picker.- GPT Image 2.5 Flare and Grok Imagine Video still create your media. They power the Image generation and Video generation tools, which work with every chat model.
- Other media models are no longer available. These include GPT Image 2, Grok Imagine Image, Nano Banana 2, Nano Banana 2 Lite, FLUX Kontext Pro, FLUX Kontext Max, Seedance 2.5, Veo 3.1, Kling v3.0, and Flux 3.
- Your saved media is unchanged. Images and videos you already created stay in their chats and in Library.
- Controls are narrower than some former models. For example, videos are now up to 10 seconds, and images come in three shapes. Review the result before you rely on it.
Wait for longer media runs
Image and video generation can take longer than short chat replies. If Chat completion notifications are on, Babbily can notify you when an image or video finishes while the tab is inactive or in the background. Turn notifications on or off in Settings > Account with the Chat completion notifications switch.Prompt tips
- Describe what should be in the frame.
- State the style and mood.
- Say whether you want a square, portrait, or landscape image.
- For video, describe motion and pacing.
- For edits, describe only what should change and what should stay the same.
- Keep brand-sensitive or regulated imagery specific and review outputs carefully.
Usage impact
Image and video generation usually use more of your API usage budget than short text replies. If you are near your limit, check the usage card before generating media.Media recipe
Follow a step-by-step workflow for creating images or videos from prompts and supported source media.