Your AI image studio
One clear visual, from direction to finished image
Beatra AI Image Studio creates images from a written brief, composes them from one to four ordered reference images, or changes only the part of an existing image you name. It returns up to four PNG images at 1K, 2K, or 4K for product photos, ad creative, brand visuals, posters, social graphics, illustrations, and concept art. Beatra AI Image Studio works in Claude Code, Codex, Cursor, and other agents that run Skills.
Install Beatra AI Image Studio and tell me when I need to sign in: https://beatra.ai/skills/beatra-ai-image-studio/install.md
For agents:Install guideMarkdown version
Send it to WorkBuddy, Codex, Claude Code, or another agent to install. Sign in once the first time you use it.GitHub ↗Docs ↗Use it on your phone: WeChat mini program
Three useful ways to create
Direction before generation
Define the message, subject, composition, style, light, color, destination, and must-keeps before creating.
Start from the source that matters
Use a written brief, one to four ordered reference images, or one base image while keeping your chosen starting point clear.
Review before another version
Inspect the generated image, then choose one focused edit, new composition, or new generation.
How it is made
Every image starts with a clear visual brief
Confirm the purpose, message, subject, sources, composition, style, light, canvas, output count, and review criteria before creation.
Define one message and its destination · Choose subject, composition, style, light, and color · Inspect every source image before using it · Review each generated image against the brief
Generate, compose, or edit
Choose the starting point that matches your real material and intended change.
Generate a new image from a written direction · Compose a new image from ordered references · Edit an existing base image · Use a reviewed result as the source for one focused follow-up
Choose a model that fits this image
Available choices depend on the source images, canvas, number of results, relationship between results, and requested creative settings. A named compatible model remains your choice.
Match the model to the requested image and sources · Keep a named compatible choice · Confirm model and input compatibility before creation · Review the actual result against the visual brief
Watch the image before choosing the next step
Check message, subject, composition, style, text, actual dimensions, and destination fit, then change only what matters.
Compare each result with the visual brief · Check actual dimensions and destination fit · Inspect every available option · Choose one focused next step
Images for real production needs
Create product images
Place a supplied product into a new scene or generate a new product concept, then check shape, color, label, and destination fit.
Build ad and brand visuals
Turn one message, palette, and audience into a campaign image or brand-led visual draft.
Make social graphics and posters
Plan one image around the channel, orientation, room for text, and message.
Develop illustrations and concepts
Explore a named visual style for an illustration, cover, scene, or concept image.
Change a photo or background
Use an existing image as the base and focus the requested change on an object, region, background, or overall treatment, then inspect the whole result.
Compose from reference images
Use ordered product, subject, style, or scene references to guide a new image, then check what carried through.
Inputs and outputs
Input · Text — Written image brief
Input · Image — Ordered references or a base image to edit (optional)
- Formats
- JPG, PNG, WEBP
- Specifications
- up to 4 images per request
Output · Image — Generated, composed, or edited images
- Formats
- PNG
- Specifications
- up to 4 images per request; 1K, 2K, 4K; editable
| Input or output | Type | Formats | Specifications |
|---|---|---|---|
| Input | Text — Written image brief | — | — |
| Input | Image — Ordered references or a base image to edit (optional) | JPG, PNG, WEBP | up to 4 images per request |
| Output | Image — Generated, composed, or edited images | PNG | up to 4 images per request; 1K, 2K, 4K; editable |
Specifications last checked · Sources: Model catalog · Documentation · Terms of Service
Questions
What can Beatra AI Image Studio make?
It can create product images, ad creative, brand visuals, posters, social graphics, illustrations, concept art, photo variations, and background changes from text, references, or an existing base image.
Can it use my own photos or product images?
Yes. One to four supplied images can guide a new composition, or the first supplied image can remain the base for an edit.
Can it change only part of an image?
Yes. A focused edit changes only a defined area of an existing base image.
Which image models does it use?
It matches the image goal, reference materials, canvas, output count, and creative controls with current models, and keeps a named model when it fits those inputs.
How is a series made to match?
Every image in the series reuses the same direction, wording, reference images, subject cues, style, logo placement, and text plan.
Related Skills
