Can MagicPost generate visuals and carousels?

Yes. MagicPost generates both LinkedIn carousels (multi-slide PDF documents) and single-image visuals with AI, from a topic, a URL, a file, or one of your posts.

Both live on the same page, so you never leave MagicPost to design the image that goes with your post.

Two tools, one Media page

Everything happens on the Media page. It has a Library for the media you keep, plus a Visuals tab, a Carousels tab, and GIFs.

Both features are in Beta. They work, and you can publish what they produce, but expect them to keep changing.

The two tools solve two different problems:

Carousels. A multi-slide deck of 1–20 slides, in a fixed 1080×1350, 4:5 format. Exported as a PDF, one page per slide, and published on LinkedIn as a document.

Visuals. A single image, in 4:5 (1080×1350), 1:1 (1080×1080), or 1.91:1 (1200×627). Exported as a JPEG and attached to a post on LinkedIn.

In short: a carousel is a document your readers swipe through, a visual is a single image in the ratio you pick.

Start a carousel

To start a carousel, open the Carousels tab and click New carousel. You can Generate with AI, Start from scratch, or begin From a template. There are 32 templates and 12 design styles to build from. See how to generate a LinkedIn carousel with AI for the full flow.

Two ways to make a visual

A visual can be built two ways, and they answer different needs.

Customizable models. Pick one of 18 structured models across 5 families (Quote, Short post, Data & numbers, Diagram, Lists & tables), then edit it on a full canvas: text, colors, layout, everything. This is the route when you want control. See how to create a visual from a customizable model.

AI visuals and AI mockups. The other two types, AI visuals and AI mockups, are driven by an instruction instead of a canvas. You get more than 30 models and about 20 rendering styles, and the AI produces the image. See what AI visuals and AI mockups are.

For AI images, you write an Instruction (up to 500 characters) and can ask for up to 2 variants at once. A version restore or a layout switch is free. Only a fresh generation under a different design counts.

What both share

Carousels and visuals share the same building blocks, so once you learn one, the other feels familiar.

Four sources. Every generation starts from One of my posts, a Topic, a URL, or a File. Point MagicPost at what you have, and it drafts from there. Carousels accept source files in PDF, .docx, .txt, or .md, up to 50 MB.

A shared brand kit. One brand kit, one identity plus 1–3 styles, applies to both carousels and visuals. The identity holds your name, professional subtitle, profile photo, and logo, each of which you can toggle on or off. See how to set up my brand kit.

The same way to publish. When a carousel or visual is ready, use Add to a post to drop it into a new or existing post. See how to add a carousel or visual to a LinkedIn post.

An AI copilot. Both editors have a copilot: type a Request a change and MagicPost reworks the design in place. On a visual you might ask it to Make the text more impactful; on a carousel, Make the hook more impactful.

Separate quotas. Generations run against a monthly allowance on a rolling 30-day window, counted per feature. The free plan is blocked behind an upgrade wall, and you get a small one-time allowance during the free trial, which does not refill. See how many visuals and carousels can I generate.

Need help?

Reach out via the in-app chat, or book a 30-minute onboarding call: https://cal.com/magicpost-team/contact-sales-team