Free tools Windows power users keep installed
One-click scans. No signup required.
To get a useful result from a reference image, tell the image model what you want to make, what the reference should contribute, and what must change or stay the same. Then inspect the result and request one focused revision at a time. A clear prompt is more useful than a long one: OpenAI Academy says one to three sentences are usually enough to communicate the image’s purpose and essential constraints.
What to include in a reference-image prompt
Start with the intended output and the visual details that matter. Depending on the task, that may include the subject, its action, the setting, composition, style, framing, or lighting. Mention only details that help define the result; adding words for their own sake does not improve the instructions.
For an edit, make the requested change explicit and separately name the features to preserve. For example, you might ask to replace a subject’s clothing while keeping the person’s identity, camera angle, and background unchanged. Do not rely on the model to infer which parts of an image are off limits.
Give every reference image a clear role
When you upload more than one image, identify each by its order and purpose. Say whether it is the scene to edit, a layout guide, or a style reference, and explain how the images should be combined. Spatial terms such as “foreground,” “background,” “left,” and “right” help clarify relationships.
For example: “Image 1 is the scene and layout reference. Image 2 is the style reference. Use Image 1’s subject placement and Image 2’s color palette; keep the text and background from Image 1 unchanged.” This communicates your intent, but it cannot guarantee that every detail will be preserved.
Use this prompt scaffold
Adapt this template to your task rather than filling every bracket mechanically:
Rank #2
Use Image 1 as the [subject, layout, or style] reference. Create [intended output and use]. Keep [required details] unchanged. Change [specific elements] to [description]. Match [composition, lighting, palette, or visual style] where relevant. Do not add [specific unwanted elements].
For multiple uploads, add a sentence assigning each image a role before describing the result. If exact text, a product shape, a person’s identity, or another detail matters, name it plainly and check it in the output.
Recommended Free Tools
Revise the result one issue at a time
- Compare the result with your requirements. Check whether the requested subject, composition, style, and details appeared, and whether anything you asked to preserve changed.
- Identify the most important miss. Avoid rewriting the entire prompt when only one element needs correction.
- Give a targeted revision. For example: “Keep the current composition and lighting, but move the title to the upper-left corner.”
- Inspect again. Confirm that the correction worked without introducing another unwanted change.
OpenAI’s image guidance recommends checking details such as text accuracy, identities, product shapes, labels, and relationships between objects. In its documented workflow, a mask can guide an edit but is not an exact boundary guarantee.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Reference controls depend on the tool
A reference image can function as a style guide, composition guide, edit target, or image-to-image input. Do not assume that one service’s controls or behavior apply to another.
| Tool or workflow | Documented reference handling | What to keep in mind |
|---|---|---|
| ChatGPT image guidance | Reference images can guide generation or editing; identify multiple images by order and explain each one’s contribution. | Use descriptive instructions, then inspect and revise the result. |
| OpenAI API image workflow | Prompts can specify subject, composition, style, and constraints; edit prompts can distinguish changes from features to preserve. | A mask guides an edit but may not follow an exact boundary. |
| Adobe Photoshop | The documented Reference image interface offers Style and Composition choices, followed by a text prompt. | Use the options available in that interface; these are product-specific controls. |
| Stability AI image-to-image API | The cited API reference takes a prompt, an input image, and a strength parameter. | The strength parameter belongs to this API workflow; it is not a universal prompt control. |
For the relevant instructions, see OpenAI Academy’s guide to creating images with ChatGPT, the OpenAI API image guide, Adobe’s Photoshop Reference image documentation, and the Stability AI image-to-image API reference.
Quick Recap
A quick checklist before generating
- Have I stated the image’s purpose and the subject or scene?
- Have I described the composition and style only as specifically as the task requires?
- Have I assigned a role to every uploaded reference?
- For an edit, have I said both what to change and what to keep fixed?
- Have I checked the tool’s actual reference-image options instead of assuming they match another product?
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.
