Hi there,

Checking a design meant switching formats one by one and hoping nothing broke along the way. The editor now has an All formats view: every format of your design, side by side on a single board.

Compare every format at a glance

Click the All formats button in the top bar, right next to the Select tool. The board shows all your formats in the same order as the Format Manager. Each one has a label with its name and size, for example facebook-post 1200×1200, so a cut-off headline or an off-balance layout stands out straight away. Zoom in and move around the board to inspect the details.

On multi-page designs, the button is called All pages and the board shows every page side by side.

Play animations in place

On an animated design, press the play button above any format. Its label turns mint green, and a progress bar with the elapsed time shows where you are in the animation. Play one format or several, without leaving the board.

A view for checking, one click from editing

The board is read-only: it is there to check your work, so nothing moves by accident. Spot something to fix? Click Select in the top bar to go back to the editor.

Tip: on a design with a 1200×1200 post, a 1080×1920 story and a 300×600 half-page, play the animation on each format and check that the headline stays readable in the narrowest one (300 px wide) before you export.

Hi there,

Two API releases this week. You can now manage designs entirely through REST, and the design import is more reliable.

Manage designs through REST

Three endpoints that were only available through MCP are now part of the REST API:

• PUT /designs/{designId}/name: rename a design.
• PUT /designs/{designId}/project: move a design to another project. This resets its approval status.
• DELETE /designs/{designId}: delete a design. It cannot be restored.

All three are in the Write rate-limit tier.

Smoother design imports (alpha)

• Skew: the import now supports skew on rounded-corner and window masks, and on frame and window shapes. Export returns it too.
• Layer names follow the editor rule (a-z, 0-9, - and _). Older names are converted, with a layer_renamed warning.
• Auto-layout groups: members are placed by the import, as the editor does when you save.
• No more endless polling: a stuck import times out after 60 min for the build and 30 min for previews.

Fixes

• Export as import now returns min_font_size, so text auto-resize survives a round trip.
• Skewed text renders with the same slant as in the editor, in images and PDFs.
• A failed import no longer leaves a half-built design.

Tip: the new endpoints make it easy to tidy up a workspace from a script, for example moving every design of a finished campaign to an archive project.

Hi there,

Framing a photo, a logo or a product shot used to mean stacking an extra shape behind it. Image layers now have a Border: set a width and a color and the frame is drawn for you.

Switch it on, set two values

Select an image layer and turn on Border in Styles (between Mask and Blur). Two settings:

• Width: set it with the slider or type a value (default 1).
• Color: pick it in the color picker (default #000000).

The border is drawn outside the image box and follows the mask, so a rounded or circle mask gets a border of the same shape.

Good to know

• Works in the editor and through the API: set stroke_width and stroke_color on an image element when you generate.
• Borders are included when you export a design with the API and when you import it back.

Tip: for a thin outline on a profile photo, try a width of 4 with a white color, then set the same values through the API to keep every generated visual consistent.

Hi there,

Ragged right edges look untidy on longer blocks of text, like descriptions or legal lines. Text layers now support justified alignment: every line stretches to the full width of the box.

Three ways to justify

In the editor, select a text layer and open Styles > Text alignment. The fourth icon (justify) opens a dropdown with three options:

• Justify, last line left
• Justify, last line centered
• Justify, last line right

Every line fills the box width, except the last line of each paragraph, which keeps the alignment you picked.

Good to know

• Works in the editor and through the API: text_align accepts justify-left, justify-center and justify-right on text layers.
• The last line of each paragraph keeps its alignment (left, centered or right).
• The existing Left, Center and Right options are unchanged.

Tip: justified text looks best in wide boxes where each line holds several words, like product descriptions or legal mentions.

Hi there,

Generating variations from a form is fast, but a different brand or market often needs a different typeface. Quick generation now lets you choose the font of a text layer, right in the form.

Change the look, not the design

As the designer, turn on Enable Font Selection for each text layer in Settings > Layers configuration > Quick generation Settings (it is off by default), then click Save changes. In the Quick generation form, that text field now shows a font dropdown with search, so you can adapt the tone of a visual in a few clicks.

Good to know

• Available in Quick generation only, not in spreadsheet generation.
• The font list offers Google Fonts.
• Font selection is enabled per text layer by the designer.

Tip: generate the same banner twice, once with a clean sans-serif for a corporate audience and once with a playful font for social, without touching the original design.

Hi there,

Adapting one print design to a flyer, a poster and a business card by hand takes time. Auto format AI now works on print designs too, so you can restructure a layout for new print formats in a single click.

Print formats, adapted by AI

Open your print design, click Add new formats, pick your target formats and turn on Ai-powered formats. Each action costs 10 AI credits, whether you create one format or several. The AI reads your layout and rearranges your layers for each new format, instead of simply stretching them.

Good to know

• Works with the print formats from A0 to A4 and with Custom sizes.
• Review each new format before sending it to your printer.

Tip: turn a portrait A4 flyer into a landscape poster in one action, then fine-tune the layout if needed.

Hi there,

Revealing a headline one letter at a time is not always what you want. The Typewriting text effect now lets you choose how your text appears and when it plays, so your animated banners match the pace of your message.

Reveal per letter, word or line

A new Split dropdown in the Animation tab lets you reveal your text per letter, per word or per line (Letter is the default). Short punchy headlines work well word by word, and multi-line captions read nicely line by line.

Play it in, out, or both

The new Timing option offers In (the default), Out or In & out. Out plays the effect in reverse, with the last unit disappearing first, so your text can leave as smoothly as it arrived.

Decorations that type with the text

• Underline, strike-through, diagonal and per-character highlight now appear together with their characters, instead of showing from the first frame.
• A paragraph background stays whole.
• Arabic and Hebrew texts now animate too, per word or per line.

For developers, the API accepts the optional textEffectSplit and textEffectTiming fields on textEffect keyframes, and the export returns them. Nothing breaks for existing integrations.

Tip: set Split to Word on a 6-word headline and use In & out, so the message builds up word by word and then clears the screen at the end of your banner.

Hi there,

Two updates for AI image generation this week: a new model built for precise edits, and reference images that now make the most of each model.

New model: Ideogram 4.5

Ideogram 4.5 is now available in the Text-to-Image model list.

Ideogram's models are known for their crisp typography, and this new version focuses on precise editing: change one part of an image, like a color, a piece of text or an element of the scene, while the rest stays untouched. Details also hold up when you edit the same image several times in a row, so you can refine a visual step by step without it drifting away from the original.

That makes it a great fit for product photos to retouch, signs or packaging text to adapt, and any visual where a small change shouldn't ruin everything else. You can add up to 5 reference images to guide it.

Select it from the model picker in the AI toolbar. As always, its credit cost is shown before you generate.

Reference images: the right number for each model

Until now, every Text-to-Image model was limited to 4 reference images, whatever it could actually handle. Many models can take a lot more, and some take fewer.

Abyssale now applies each model's real limit:

• Up to 16 images with GPT Image 2.5 Sunburst, GPT Image 2.5 Flare and GPT Image 2
• Up to 14 with Nano Banana 2, Nano Banana Pro and the Gemini models
• Up to 10 with GPT Image 1.5, 8 with FLUX.2 Pro, and 7 with Seedream
• Fewer for a few lighter models, like Nano Banana (3)

In practice, that means you can give the AI much more to work from: several product angles, your brand's visual style, a logo and a mood board, all in a single generation. The result is closer to what you have in mind, in fewer attempts.

The same limits apply when you generate through the API.

Hi there,

Abyssale is no longer English-only. The whole interface is now available in French, with German, Spanish, Portuguese (Portugal) and Portuguese (Brazil) joining in beta.

Switch in one click

Go to Settings → My Profile → Language and pick your language from the list. The entire app switches instantly: menus, panels, settings, everything.

Your choice is remembered in your browser, so Abyssale opens in your language every time you come back.

Already in the right language

You may not even need to change anything. If your browser is set to French, Abyssale now opens in French automatically. Otherwise, it opens in English, and you can switch anytime from your profile.

Five languages, more to come

French is fully available. German, Spanish, Portuguese (Portugal) and Portuguese (Brazil) are in beta: you can already use them, and we're still polishing the translations. If you spot a wording that feels off, let us know through the chat, it helps us improve them quickly.

Tip: sharing a link with a colleague who works in another language? Add ?lang=fr or ?lang=en at the end of any Abyssale URL to open it directly in French or English.

Hi there,

Aligning layers by eye works, until it doesn't. The new helper grid displays a grid over your format in the Design Builder, and your layers snap to it as you move them. Precise, balanced layouts in a few seconds.

Layers that snap into place

Turn the grid on and drag any layer: as soon as one of its edges gets close to a grid line, it snaps right onto it. Its position updates to the exact value, so every element lines up perfectly with the others.

Snapping works on each axis independently, so you can lock a layer horizontally and keep moving it freely up and down. And you can still type exact positions whenever you want.

The grid is only a guide while you design: it never shows up in your generated images, videos or PDFs.

On and off in one shortcut

Show or hide the grid from the zoom menu in the top toolbar, with Display helper grid. Or, even faster, press Ctrl + Shift + G on Windows, Cmd + Shift + G on Mac.

Hide the grid, and your layers move freely again.

Make it your own

In the zoom menu, Grid settings… lets you adjust:

• the block size, which is also the step your layers snap to
• how often thick lines appear, to mark out the main sections of your layout
• the colors of the thin and thick lines, so the grid stays visible on any background

Your settings are saved as your personal preferences and apply to every design you open, without changing the designs themselves.

Tip: pick a block size that divides your format evenly. On a 1080 × 1080 px format, 108 px gives you a clean 10 × 10 grid.