August 29, 2026AIExplore

How to Use Adobe Firefly Style and Composition References

Upload style or composition references in Adobe Firefly, adjust Strength for layout adherence, and write prompts that say what to keep versus what to change.

Adobe Firefly

Adobe Firefly style references guide look and feel. Composition references guide outline and depth structure. Official Style reference and Composition help explain uploading references, adjusting Strength for composition adherence, and pairing references with a clear text prompt. Use references when brand consistency or layout control matters more than a cold start Text to Image prompt. Overview: /explore/firefly.

This guide shows when to use style versus composition controls, how to write keep versus change language, and how to iterate Strength. Pair with Text to Image at /blog/how-to-use-adobe-firefly-for-text-to-image-generation, commercial workflows at /blog/how-to-use-adobe-firefly-for-commercial-campaigns-and-brand-visuals, and iterative prompting at /blog/how-to-improve-adobe-firefly-results-with-iterative-prompt-refinement.

When references beat a cold prompt

Use a style reference when palette, medium, or finish must match a known look. Use a composition reference when camera angle, subject placement, or depth structure must stay. Use both carefully. Conflicting references fight each other.

Step by step workflow

1. Pick style, composition, or both

Decide which problem you are solving. Style without composition leaves layout free. Composition without style leaves look free.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Reference plan Need: brand teal finish Need: centered bottle layout Style ref: yes Composition ref: yes Strength: start mid, adjust. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Reference plan Need: brand teal finish Need: centered bottle layout Style ref: yes Composition ref: yes Strength: start mid, adjust" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Inputs:
Use the actual source material required for this task, together with the intended audience, destination, format, and quality requirements. Specify relevant files, text, URLs, structured data, reference assets, dimensions, language, tone, timing, model or generation settings, credentials, field mappings, or integration values when they apply. Keep secrets out of the example and use the product's supported credential or configuration mechanism.

Workflow:
1. Start with one representative input that contains the important characteristics of the real workload.
2. Configure adobe-firefly specifically for the requested task and make the important settings explicit.
3. Run the first pass and inspect the result against the objective.
4. Correct only the settings or source material responsible for a failed requirement instead of changing everything at once.
5. Validate the intermediate result before passing it to the next step.
6. When the result meets the requirements, save the successful configuration so the same process can be repeated consistently.
7. If this is a multi-step workflow, verify each handoff and keep a clear fallback or review path for failures.

Requirements:
The example must use supported functionality only. Do not invent product features, pricing, limits, integrations, model names, API behavior, or unavailable settings. Preserve important source data and formatting. Validate required fields before processing, handle empty or invalid input explicitly, prevent accidental duplicate processing where relevant, and stop or route the item for review when a required step fails. For credentials, use the supported secure configuration rather than placing secrets directly in the content.

Expected output:
The final result should directly satisfy the "Reference plan Need: brand teal finish Need: centered bottle layout Style ref: yes Composition ref: yes Strength: start mid, adjust" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

2. Write keep versus change

State what the reference should control and what the prompt should invent. Vague prompts let the model copy unwanted props from the reference.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Prompt pairing Keep from style: watercolor paper and pastel palette Change: subject becomes bicycle at cafe Locks: no text, no logos. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Prompt pairing Keep from style: watercolor paper and pastel palette Change: subject becomes bicycle at cafe Locks: no text, no logos" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Prompt pairing Keep from style: watercolor paper and pastel palette Change: subject becomes bicycle at cafe Locks: no text, no logos" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

3. Adjust Strength and regenerate

If composition copies too much, lower Strength. If layout drifts, raise Strength slightly and simplify the prompt. Generate four options before large changes.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Strength log Mid: good layout, wrong prop copied Lower: freer layout, better subject Keeper: lower Strength + clearer subject line. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Strength log Mid: good layout, wrong prop copied Lower: freer layout, better subject Keeper: lower Strength + clearer subject line" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Strength log Mid: good layout, wrong prop copied Lower: freer layout, better subject Keeper: lower Strength + clearer subject line" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Practical style and composition prompts

Watercolor style

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Style reference: soft watercolor paper texture. Subject: bicycle leaning on a cafe wall, loose brush edges, muted pastel palette, no text. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Style reference: soft watercolor paper texture. Subject: bicycle leaning on a cafe wall, loose brush edges, muted pastel palette, no text" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Style reference: soft watercolor paper texture. Subject: bicycle leaning on a cafe wall, loose brush edges, muted pastel palette, no text" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Catalog composition

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Composition reference: centered bottle on seamless. Prompt: matte black skincare bottle, soft top light, subtle reflection, studio catalog look, no label text. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Composition reference: centered bottle on seamless. Prompt: matte black skincare bottle, soft top light, subtle reflection, studio catalog look, no label text" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Composition reference: centered bottle on seamless. Prompt: matte black skincare bottle, soft top light, subtle reflection, studio catalog look, no label text" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Brand palette style

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Style reference: brand teal and sand abstract. Subject: runners on a coastal path at dawn, naturalistic people, generous negative space, no logos. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Style reference: brand teal and sand abstract. Subject: runners on a coastal path at dawn, naturalistic people, generous negative space, no logos" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Style reference: brand teal and sand abstract. Subject: runners on a coastal path at dawn, naturalistic people, generous negative space, no logos" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Editorial layout

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Composition reference: model left third, empty right. Prompt: linen coat fashion shot, soft rooftop light, keep empty right for later type. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Composition reference: model left third, empty right. Prompt: linen coat fashion shot, soft rooftop light, keep empty right for later type" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Composition reference: model left third, empty right. Prompt: linen coat fashion shot, soft rooftop light, keep empty right for later type" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Illustration style

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Style reference: flat vector poster look. Subject: friendly delivery bike, bold shapes, limited colors, no trademarks. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Style reference: flat vector poster look. Subject: friendly delivery bike, bold shapes, limited colors, no trademarks" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Style reference: flat vector poster look. Subject: friendly delivery bike, bold shapes, limited colors, no trademarks" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Food overhead composition

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Composition reference: plate centered overhead. Prompt: avocado toast on stoneware, bright window light, herb garnish, keep plate center. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Composition reference: plate centered overhead. Prompt: avocado toast on stoneware, bright window light, herb garnish, keep plate center" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Composition reference: plate centered overhead. Prompt: avocado toast on stoneware, bright window light, herb garnish, keep plate center" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Film still style

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Style reference: muted film still grain. Subject: rainy city street at night, neon reflections, cinematic shallow depth of field. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Style reference: muted film still grain. Subject: rainy city street at night, neon reflections, cinematic shallow depth of field" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Style reference: muted film still grain. Subject: rainy city street at night, neon reflections, cinematic shallow depth of field" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Product three quarter

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Composition reference: three quarter product angle. Prompt: white sneakers on concrete, soft daylight, subtle scuffs, same camera height. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Composition reference: three quarter product angle. Prompt: white sneakers on concrete, soft daylight, subtle scuffs, same camera height" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Composition reference: three quarter product angle. Prompt: white sneakers on concrete, soft daylight, subtle scuffs, same camera height" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Kids book style

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Style reference: soft crayon kids book. Subject: child reading under a blanket fort, warm lamp glow, gentle edges. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Style reference: soft crayon kids book. Subject: child reading under a blanket fort, warm lamp glow, gentle edges" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Style reference: soft crayon kids book. Subject: child reading under a blanket fort, warm lamp glow, gentle edges" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Architecture depth

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Composition reference: hallway vanishing point. Prompt: bright modern corridor, oak floors, soft daylight, keep strong depth. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Composition reference: hallway vanishing point. Prompt: bright modern corridor, oak floors, soft daylight, keep strong depth" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Composition reference: hallway vanishing point. Prompt: bright modern corridor, oak floors, soft daylight, keep strong depth" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

More copyable examples

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Textile style Style reference: woven linen close up. Subject: throw pillow on sofa, soft folds, muted sand, lifestyle light. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Textile style Style reference: woven linen close up. Subject: throw pillow on sofa, soft folds, muted sand, lifestyle light" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Textile style Style reference: woven linen close up. Subject: throw pillow on sofa, soft folds, muted sand, lifestyle light" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Poster negative space Composition reference: large empty top band. Prompt: quiet meadow with soft fog, hopeful mood, reserve top for title later. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Poster negative space Composition reference: large empty top band. Prompt: quiet meadow with soft fog, hopeful mood, reserve top for title later" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Poster negative space Composition reference: large empty top band. Prompt: quiet meadow with soft fog, hopeful mood, reserve top for title later" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Jewelry macro style Style reference: high key jewelry catalog. Subject: gold ring on marble, soft speculars, clean reflections, no logo. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Jewelry macro style Style reference: high key jewelry catalog. Subject: gold ring on marble, soft speculars, clean reflections, no logo" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Jewelry macro style Style reference: high key jewelry catalog. Subject: gold ring on marble, soft speculars, clean reflections, no logo" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Street photo style Style reference: candid street photography. Subject: friends sharing coffee outdoors, noon light, natural color, no posed stiffness. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Street photo style Style reference: candid street photography. Subject: friends sharing coffee outdoors, noon light, natural color, no posed stiffness" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Street photo style Style reference: candid street photography. Subject: friends sharing coffee outdoors, noon light, natural color, no posed stiffness" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Pack shot composition Composition reference: bottle and box side by side. Prompt: cream skincare set, soft top light, blank labels, catalog spacing. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Pack shot composition Composition reference: bottle and box side by side. Prompt: cream skincare set, soft top light, blank labels, catalog spacing" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Pack shot composition Composition reference: bottle and box side by side. Prompt: cream skincare set, soft top light, blank labels, catalog spacing" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Painterly landscape Style reference: loose impressionist brush. Subject: coastal cliff path at golden hour, gentle haze, no text. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Painterly landscape Style reference: loose impressionist brush. Subject: coastal cliff path at golden hour, gentle haze, no text" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Painterly landscape Style reference: loose impressionist brush. Subject: coastal cliff path at golden hour, gentle haze, no text" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: UI mock mood Style reference: soft gradient product site hero. Subject: abstract device silhouette on pastel gradient, empty space for UI later, no logos. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "UI mock mood Style reference: soft gradient product site hero. Subject: abstract device silhouette on pastel gradient, empty space for UI later, no logos" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "UI mock mood Style reference: soft gradient product site hero. Subject: abstract device silhouette on pastel gradient, empty space for UI later, no logos" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Scenario:
A realistic adobe-firefly workflow is being prepared for Use Adobe Firefly for Text to Image Generation using adobe-firefly. The starting requirement is: Seasonal campaign style Style reference: previous winter campaign still. Subject: scarf and mug on windowsill, cool daylight, cozy mood, no trademarks. The work should be tested on a representative input first so the result can be reviewed before the same process is applied to the full project or production workload.

Objective:
Create a production-ready result for "Seasonal campaign style Style reference: previous winter campaign still. Subject: scarf and mug on windowsill, cool daylight, cozy mood, no trademarks" that directly supports the article use case. The output should be specific, reviewable, and reproducible rather than a generic demonstration. Keep the requested goal as the primary outcome and avoid adding unrelated features, assumptions, or unsupported capabilities.

Expected output:
The final result should directly satisfy the "Seasonal campaign style Style reference: previous winter campaign still. Subject: scarf and mug on windowsill, cool daylight, cozy mood, no trademarks" requirement and be understandable without guessing what was supplied or configured. A successful run should produce the intended output in the requested format, with the important values and workflow decisions clear enough to reproduce the result. If the workflow can fail, the expected behavior should also make the failure visible and provide a clear next action instead of silently producing an incomplete result.

Tips

One clear job per reference type
Say what must differ from the reference
Lower composition Strength when unwanted layout details stick
Keep prompts descriptive, not command based
Generate four options before changing Strength and prompt together
Confirm commercial use and beta labels for your outputs

Verification notes

Adobe Style reference help covers look and feel guidance. Composition help covers outline and depth structure with Strength. Confirm current control names in your Firefly UI. When credits, plan features, or model access differ from memory, open Adobe Firefly plans and the generative credits FAQ. Limits change, so do not promise client deliverables from outdated screenshots.

Keep a short creation log beside your Firefly history: prompt, aspect ratio, reference used, and the one field you will change next. That habit makes four option comparisons useful instead of random.

When credits, Free daily limits, or premium feature access look different than yesterday, open Adobe Firefly plans and your account generative credit indicator. Plan details change, so do not rely on memory for client promises.

Treat the first Generate as a sketch. Review all four options before rewriting the whole prompt. Small descriptor edits teach more than full rewrites.

If you collaborate, store winning prompts and reference files on a shared brand page. Reuse the skeleton and change subject or channel crop only.

For social clips and paid placements, decide aspect ratio before you generate. Asking for a new crop after the fact wastes time and may cost credits on premium paths.

Separate exploration from release prep. Explore freely, then re check Adobe commercial use wording, beta labels, and Content Credentials needs before monetizing.

Close each session by naming keepers and archiving rejects. A clean library saves time when a stakeholder asks for the mug shot from Tuesday.

Prefer clear visual language over vague vibes alone. Subject, light, materials, and one mood word outperform a pile of adjectives.

Move brand critical finishing into Photoshop or Illustrator when edges, type, or logo placement must be exact. Firefly accelerates exploration. Design tools finish production.

Confirm whether your current feature shows a beta label before you claim commercial clearance. Adobe commercial wording applies to features without the beta label per product materials.

A sharp, well lit reference outperforms a busy collage. Garbage in still produces drift even with good Strength settings.

When style and composition fight, run two passes: lock composition first with a neutral prompt, then apply style on a second generation if your workflow allows.

Document which reference file produced the keeper. Brand teams need reproducibility more than one lucky frame.

Curate a small approved reference set per brand: one style board, one pack shot layout, one lifestyle layout. Too many references create committee drift. Fewer stronger references create recognizable campaigns.

When a reference includes logos or people you cannot clear, crop those regions out of the reference file before upload if your workflow allows. Do not teach the model marks you cannot ship.

Composition Strength is a dial, not a promise of pixel identity. Treat mid Strength as the default for brand packs. Raise it only when placement errors keep repeating across four options.

Style references shine for illustration and texture continuity across a series. Composition references shine for catalog grids. Mixing both is powerful for product heroes that must match a prior campaign still.

Revisit references every quarter. Seasonal campaigns change. An outdated winter reference will fight a spring brief even when Strength and prompts are carefully written.

Common mistakes

Uploading a reference but not stating what should differ
Mixing conflicting style references in one pass
Expecting exact trademarked logos from references or text
Max Strength on composition when you only needed a soft guide
Changing Strength and the full prompt at the same time
Using a low quality reference and blaming the model

Related Adobe Firefly articles: /blog/how-to-use-adobe-firefly-for-text-to-image-generation, /blog/how-to-use-adobe-firefly-for-commercial-campaigns-and-brand-visuals, and /blog/how-to-improve-adobe-firefly-results-with-iterative-prompt-refinement. Return to /explore/firefly for the broader guide set.