Image model · Ideogram ·
Ideogram 4.5 is an image model by Ideogram. In Endless it costs from 3 credits per image and is available on every plan.
Ideogram 4.5 is Ideogram's image model for designs with readable text and for edits that change only what you ask, round after round.
Generated in Endless. Open any image to see and copy the prompt that made it.
Ideogram 4.5 is built around precise editing. You attach an image, describe the change, and the model keeps the rest of the picture in place. Pixel shifts and color drift do not pile up as you edit again, so you can refine one image across several turns. It also keeps Ideogram's long-standing strength: text inside the image that reads correctly.
Teams use it for posters, logos, social graphics and ad variations where the words matter as much as the picture. A common flow is to generate a layout, then swap a headline, recolor a product or remove a background element without redrawing everything else. In Endless you can use it in chat and on the canvas, and switch to another image model in the same workspace when a job calls for it.
Ideogram presents 4.5 as an editing model first. Its pitch is that most image models add artifacts, pixel shifts and color changes with each edit, and that 4.5 avoids that buildup, so you can refine one image over many rounds. It still generates from a prompt alone, and Ideogram describes its models as known for prompt fidelity and crisp typography. Output is native 2K.
The model accepts long natural-language prompts, up to 10,000 characters, and follows a specific prompt closely. So the order of the details matters. Runware's guide suggests the same order every time: what the image is for, then the subject, the setting, where things sit, the light and the camera.
[Deliverable and format]: [subject]. [Setting].
[Where the subject sits and what the empty space looks like].
[Light]. [Camera and lens].
[Exact copy in quotes, each line with a size, style and place].
Lead with the deliverable. "Commercial product photograph for a 16:9 web banner" tells the model how to compose before it reads a single detail. Then plan the empty space: most ads get copy laid over them later, so say which third the subject sits in, how much of the frame stays empty, and what the empty part looks like ("a plain warm white wall with nothing on it"). Name the format in the prompt as well as in the aspect ratio setting, so a 9:16 story is composed head to toe and a banner spreads across its width.
Legible text is the reason many people pick Ideogram. To get it right:
Attach the image you want to change and describe the change. Ideogram says pixels the edit does not touch are kept from your image, and the result comes back at your image's size.
You can also attach up to 4 extra reference images to guide the edit with a product, a material or a style. Say what to take from each one: the shape from one, the fabric from another.
In your workspace, open Image and pick Ideogram 4.5 in the model picker.
Write a prompt, and add reference images when you want to steer the subject, style or layout.
Generate, then refine in chat or on a canvas. Every result is saved to your workspace assets.
by Google
Hard briefs that need fine detail, accurate text and many references
From 20 credits per image
by Google
Design-ready images and edits that keep your subject consistent
From 9 credits per image
by Google
Fast everyday generation and edits with references, up to 4K
From 5 credits per image
Ideogram 4.5 next to other models: price, maker and plan side by side.
Start free, create a project, and bring your brand’s references together. Or book a demo to explore Endless with your team.