Grok Bot images and video: how to ask, what a clip can be, and where the results land

Disclosure: aitamer.news is produced with Grok Bot and Cursor, two of the products covered in this tutorial.

A Bot can make pictures and short video clips when you ask. You describe what you want in the chat, as you would any other task, and the result comes back as a message. This tutorial covers how to ask, what an image or a clip can be, where the finished files go, and the reasons a Bot sometimes answers with something other than what you asked for. The page behind all of it is the Cursor help page on creating images and videos, read on 29 October 2026. SpaceXSI’s files and results page now carries the same limits. If a menu name below disagrees with the app, follow the app.

Ask for it in words

Tell the Bot what you want: the subject, the style, and the mood, in your own words. The help page is clear that a Bot creates an image only when you ask for one. It will not decorate a reply with pictures on its own.

To build on pictures you already have, share them in the chat and say how to use them. The Bot treats them as references for the new image. Episode 16 covers attaching files and the size limits.

An image can be square, wide, or tall. The help page lists five shapes: 1:1, 4:3, 3:4, 16:9, and 9:16. Name the one you need in the request, for example “16:9 for a slide” or “9:16 for a phone story”.

Here is a request that gives a Bot what it needs:

Make a 16:9 image of a quiet harbour at dawn, cut-paper style,
muted blues with one warm orange boat. No text in the image.

That template is this desk’s, not the help page’s. It works because it answers the questions the page says matter: what, in what style, in what mood, and in what shape.

When to ask for a page instead

Image creation is not used for charts, plots, or tables. For those, the help page sends you to a page: something the Bot builds for you to read, with text, tables, charts, and controls such as filters. This desk’s reason to follow that: a chart drawn as a picture can look right and still be wrong, while a page keeps the numbers as numbers.

The same rule of thumb works for anything with exact words in it. If the result must say something precisely, ask for a page or a document and add the picture beside it.

Four ways to start a clip

Ask for a video, an animation, or a clip. The help page lists four starting points:

1. A description of the scene. Words only.
2. One image you share. It becomes the first frame, and the clip moves on from there.
3. Two images you share, one for the start and one for the end. Use the same image for both to make a loop.
4. Up to 14 reference images showing how people, objects, clothing, or settings should look.

On top of any of these, you can share up to four images that must appear at set moments inside the clip.

A clip is not instant. The page says creating a video can take a few minutes, so ask, then keep working in the chat.

What a clip can be

The help page sets these limits:

• Length: 1 to 15 seconds. The default is 6 seconds.
• Resolution: 480p unless you ask for 720p.
• Shape: 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, or 2:3. A clip made from your image keeps that image’s shape unless you ask for another.
• Images you share: JPEG, PNG, or WebP, up to about 35 MB in total for one clip.
• File: each clip is saved as an MP4.

Two of these are worth saying out loud in every request. The default is 480p, so ask for 720p when the clip will be watched on a large screen. The default is also 6 seconds, so name the length when you need a different one.

What a clip is good for

The page is direct about style. Videos look filmed. They suit a live scene, a still photo brought to life, or a character from your reference images.

They are not made for motion graphics, title cards, slideshows, product or app demos, charts, or anything that needs exact text, timing, or brand assets. If you need a logo to land on the beat or a caption spelled right, a filmed clip is the wrong tool. Make the clip, then add the words in an editor you control.

Where the results go

The Bot sends each image and video to the chat. Images it made still load there while its computer is asleep.

Everything it made is also kept in one list:

• On desktop: click the Bot’s name at the top of the chat to open its details, then open the Library tab. Images and videos the Bot sent are listed there, newest first. Images you uploaded yourself are not included. Click a row to open it full screen.
• On the phone: tap the Bot’s name at the top of the chat to open its profile, then tap Library. Tap an image or video to open it.

Save, copy, and share

On desktop, click an image or video in the chat to open it full screen. Click an image to zoom in. The left and right arrow keys move between items, the download button at the top saves the item, and Escape closes the viewer. The video player has play and pause, a seek bar, and speeds of 2×, 1.5×, 1×, and 0.5×.

Right-click works too. On an image you get Copy image and Save image…, and on a video Save video…. For the whole message, hover over it, click the three dots, and choose Download, or Download all when the message has several items.

On the phone, tap an image to open it, then use the viewer’s menu for Forward, Share, and Save. Touch and hold an image in the chat for Copy Image and Save to Photos. Tap a video to play it, and touch and hold it for Save to Photos.

Work a Bot hands off can make media too. When your Bot hands work to a task, that task can create images and videos, and your Bot sends them to the chat.

When the result is not what you asked for

The help page lists four reasons.

• You asked to see something real. To show you what a real person, place, or product looks like, a Bot finds a real image instead of making one.
• A content rule blocked it. The Bot tells you when that happens. Naming trademarked characters, brands, or real people often causes a block. Describe the subject in your own words and try again.
• Image creation is busy. The Bot may ask you to try again in a moment.
• Creation is off for your account. If the Bot says image or video generation is not available for you, it is turned off. On a team, an admin’s model settings or the team’s privacy settings can cause this. Ask your admin.

The first reason is a feature. Ask what a landmark looks like and you get a photo of it, not an invention.

What it costs

The create-media page gives no price per image or per clip, and it points to the plans page for usage. The models help page adds that Grok Bot does not show which model handled a request, and usage views record it as Grok Bot usage. Episode 06 covers how usage is counted. No page puts a number on a clip, so check Settings → Usage & Billing after a few clips to learn your own rate.

A pass you can do today

1. Ask your Bot for one 16:9 image in a style you like, with “no text in the image” in the request.
2. Share that image back and ask for a 6-second clip that starts from it.
3. Ask for the same clip again at 720p and compare the two.
4. Open the Bot’s Library and find both clips, newest first.
5. Save the one you keep, then ask for a page, not an image, the next time you need a chart.

After that, a picture or a short clip is one sentence away, and you know which jobs to give a page instead.