Concept combinations
Combine familiar subjects into coherent surreal concepts.
Create with Doubao Seedream 5.0 Lite using the modes and parameters currently available in the workspace.
These third-party visual references were supplied by users for this page. They illustrate visual directions; their origin as Seedream 5.0 Lite outputs has not been confirmed. The five prompt examples describe visible content and are not the original creator prompts.
Doubao Seedream 5.0 Lite is suited to product visuals, characters, environments, campaign assets, social content, and early creative exploration.
Generated content will appear here
Facts last reviewed:
Seedream 5.0 Lite is available here as an online image creation workflow. Use a written brief and, where enabled, authorized reference images to explore photographic, illustrative, character, layout, and concept directions.
The gallery below contains user-supplied third-party visual references. Five selected images include reconstructed prompts based on their visible subject, composition, materials, lighting, and style; the remaining examples are presented as visual references without invented prompt text.
Use concrete descriptions of subject, composition, material, light, text, and exclusions, then judge the output against your intended use.
Combine familiar subjects into coherent surreal concepts.
Explore expressive mascots, stylized figures, and polished 3D characters.
Move between watercolor, ink, charcoal, colored pencil, and bright storybook rendering.
Describe skin, metal, glass, jade, fabric, and other material cues.
Plan maps, character sheets, collages, and other structured vertical designs.
Use only images you are authorized to upload and state what each reference controls.
These examples demonstrate visual range and do not claim to reproduce the original creator prompt.
These third-party visual references were supplied by users for this page. They illustrate visual directions; their origin as Seedream 5.0 Lite outputs has not been confirmed. The five prompt examples describe visible content and are not the original creator prompts.
Third-party visual referenceColored-pencil hatching, loose contours, and layered figures create an editorial ensemble scene.
Third-party visual referenceGraphic facial shapes, ceremonial headwear, and soft volumetric light define a stylized celestial figure.
Third-party visual referenceMinimal black, red, and green brushwork captures motion with only a few confident strokes.
Third-party visual referenceEnergetic charcoal marks balance anatomical detail with a loose expressive silhouette.
Third-party visual referenceFine fur rendering and a restrained pastel gradient create a polished 3D character portrait.
These five prompts were reconstructed from the supplied images. Copy one into the workspace, then adjust names, copy, palette, ratio, and constraints.

Specify the route, landmarks, labels, paper texture, and illustration hierarchy.
Create a vertical hand-drawn Chinese travel map titled “南昌出发·环线旅行地图” on warm textured ivory paper. Draw a dotted loop route connecting 南昌、南京、日照、青岛、滕州 and 合肥. Add recognizable illustrated landmarks, coastline, lighthouse, mountains, trees, local food, tiny traveler characters, orange location dots, and short handwritten Chinese notes. Use soft watercolor, colored pencil, brown ink lettering, generous spacing, and an editorial travel-journal feel. Keep every place name clear and correctly written.

Define turnaround views, proportions, palette, facial detail, and fabric close-ups.
Design a clean vertical 3D character reference sheet on a white background for a stocky young martial-arts apprentice. Show front, side, and back turnaround views, a face close-up, a full-body proportion guide, two color swatches, and four material-detail panels. The child has short dark hair tied with coarse rope, freckles, a calm stubborn expression, a pale grey-green rough-linen wrap robe with frayed sleeves, a simple cloth belt, dark cropped trousers, and soft black shoes. Use polished animated-film character rendering, consistent proportions, neutral studio light, and concise Chinese labels.

Describe material depth, carved patterns, museum lighting, and restrained composition.
Create a vertical museum-style product study of two ancient Chinese jade ritual ornaments floating one above the other against a deep teal mineral-mist background. Use translucent pale celadon jade with frosted edges, subtle internal clouding, faint weathering, tiny gold mineral traces, and precisely carved symmetrical taotie and cloud-scroll motifs. Add soft directional light from the upper left, realistic subsurface scattering, long diffused shadows, generous negative space, and no text.

Prioritize believable pores, small imperfections, soft makeup, and unretouched light.
Create an extreme close-up beauty portrait of a young East Asian adult in warm late-afternoon window light. Frame both brown eyes, nose, cheeks, and natural lips; show fine pores, peach fuzz, subtle freckles, a few small blemishes, softly hydrated skin, minimal makeup, and loose black fringe across the forehead. Use an 85mm macro portrait look, shallow depth of field, neutral color grading, realistic catchlights, and an honest unretouched editorial finish without plastic smoothing.

Build a layered editorial composition from paper scraps, stamps, handwriting, and a cutout figure.
Create a vertical analog travel-journal collage on aged cream paper. Place a black-and-white cutout of a bespectacled East Asian man with a backpack walking from left to right across the lower half. Layer torn kraft paper, ruled notebook pages, handwritten notes, postal stamps, vintage envelopes, barcodes, paper clips, ticket fragments, red and blue airmail borders, translucent tape, and one small blue paper accent behind him. Use tactile paper fibers, imperfect cut edges, photocopy grain, warm sepia tones, dense but balanced composition, and fictional unreadable ephemera rather than real personal data.
Start with what you need to make and where it will be used.
Specify framing, palette, light, material, typography, and details to preserve.
Confirm the active mode, ratio, quality, references, and estimated points.
Check anatomy, text, identity, artifacts, and small production details before use.
Reconstructed prompts describe visible image characteristics; they are not the original prompts and may not recreate the images exactly.
AI images can contain incorrect text, anatomy, object relationships, or unintended visual similarities. Review every result and use only source material you have permission to upload.
The official link documents the provider workflow; this page describes Nano Banana Pro examples and current platform behavior.
Load a prompt, adjust the visual brief, and confirm the live settings before generating.