Recommended Free Tools
Outdated Drivers Are Slowing You Down
Free scan · exact hardware match
Windows Errors? Fix Them Before They Spread
Free scan · no reinstall
Crashes, No Sound, or Screen Glitches?
Free scan · under a minute
ThatPainter is reader-supported. When you buy through links on our site, we may earn an affiliate commission. Learn More
Ideogram’s Describe feature turns a reference image into text that you can edit and use as a prompt. In the current v4 developer API, it can return a structured JSON prompt for reuse with v4 image-generation endpoints; the original consumer feature was presented as a captioning assistant for creating and adjusting prompts.
Describe is not the same as Remix: Describe produces text from an image, while Remix reimagines the image and is intended to preserve its composition more directly. The exact options depend on whether you use the consumer app, the v4 API, a legacy API endpoint, or a documented integration workflow.
What Ideogram Describe does
Ideogram introduced Describe in April 2024 as a captioning assistant that turns images into detailed descriptions. Its announcement presented those descriptions as prompts for generating images similar to a reference, with the option to adjust the description and aspect ratio.
#1 Best Overall
The launch announcement described two ways to provide an image: choose one from the Ideogram feed or upload one from a device. That is launch-era information, not confirmation that today’s consumer-app layout, access, or plan rules are unchanged.
In the current v4 developer API, Describe can return a structured JSON prompt designed to be passed to Ideogram’s v4 generation endpoints. The response can include bounding boxes that help retain the source layout. A request can set include_bbox to false to omit them, allowing the sampler to position elements more freely.
How to use Describe
The exact steps vary by product surface. For the v4 API, the documented workflow is to provide an image to the Describe endpoint, receive a structured prompt, and use that result with a v4 generation endpoint. The API documentation lists JPEG, WebP, and PNG inputs, with a maximum image size of 25 MB; these specifications should not be assumed to apply to every consumer-app upload path.
The Tool Desk
Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →- Choose a reference image you are permitted to use.
- Submit it to the Describe feature or API endpoint available in your workflow.
- Review the returned description or structured JSON prompt.
- Edit the subject, style, layout, or other details to fit your goal.
- For the v4 API workflow, pass the resulting JSON prompt to a v4 generation endpoint. If you want the sampler to position elements more freely, the API allows bounding boxes to be omitted.
Descriptions are a starting point, not a guarantee of an exact reproduction. Check the result for details that matter to your intended image before generating.
Describe vs. Remix
Ideogram distinguishes the features by what they produce and how they use the source image.
| Feature | What it produces | How it treats the source | Best fit |
|---|---|---|---|
| Describe | Text or, in the v4 API, a structured JSON prompt | Turns visual content into editable prompt information | When you want to inspect or edit a prompt before generating |
| Remix | A reimagined image | Preserves the source composition more directly | When you want to transform an image while retaining more of its composition |
Ideogram’s Styles page also describes creating a style from up to three reference images. That is a separate workflow; do not assume it has identical controls or availability to Describe.
Describe API versions
Ideogram documents both a structured v4 Describe endpoint and a legacy Describe endpoint. They differ in output format and documented model behavior.
| Endpoint | Output | Documented details |
|---|---|---|
| v4 Describe | Structured JSON prompt | Designed for reuse with v4 generation endpoints; bounding boxes are included by default and can be omitted |
| Legacy Describe | Text descriptions | Lists V2, V3, and V4 model choices, with V3 as the documented default |
Do not confuse the legacy endpoint’s V3 default with the separate v4 structured-prompt endpoint. Both API documents list JPEG, PNG, and WebP inputs and a 25 MB maximum image size.
Other documented workflows
Ideogram’s June 2026 MCP guide demonstrates using image descriptions to identify attributes such as medium, palette, lighting, composition, grain, and mark-making, then combining those attributes into a reusable style recipe for new subjects. This is an example workflow using MCP tools, not evidence that every Ideogram user has the same integration or interface.
Ideogram’s Extend documentation also mentions using Describe when an image-editing prompt or description is unavailable, then continuing the workflow in Canvas. Separately, Ideogram describes Magic Prompt as a way to turn natural-language prompts into structured JSON that can be edited. These are related prompting workflows, not proof that every Describe output or generated image will behave consistently.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.FAQ
What is the difference between Describe and Remix in Ideogram?
Describe turns an image into text or a structured prompt that you can edit. Remix creates a reimagined image and is intended to preserve the source composition more directly.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsCan I use Describe with images I upload, or only images generated in Ideogram?
Ideogram’s launch announcement said users could select an image from the Ideogram feed or upload one from a device. The v4 API also accepts image inputs. Those sources do not establish that every current consumer-app surface has the same upload options or limits.
What image formats and size does the API support?
The v4 and legacy Describe API documentation lists JPEG, WebP, and PNG inputs, up to 25 MB per image. Treat these as API specifications, not as a guarantee of identical limits in the consumer app.
Can Describe preserve an image’s layout?
The v4 API includes bounding boxes by default, which can help retain source layout. You can set include_bbox to false to drop them and let the sampler position elements more freely. This is an API option, not a guarantee of an exact match.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
