Seedream 5 Lite
A ByteDance image model for reference-heavy edits, product preservation, relighting, outpainting, and text-aware composition.
Основные сведения
| Разработчик | ByteDance |
|---|---|
| Доступность в Banana Pie | Доступно |
| Стоимость от | 20 кредитов / изображение |
| Референсные изображения | До 8 |
Seedream 5 Lite is positioned for image workflows that combine several references with targeted changes, especially product, identity, garment, lighting, layout, and canvas edits. Its scenario coverage is broad, but the supplied evidence contains only one selected run per task, so it does not prove consistency or an advantage over other models.
Сильные стороны
- High reference allowance
The model accepts up to 8 reference images, giving creators room to supply several objects, identities, garments, or visual guides in one workflow.
- Built around asset coordination
The evidence includes selected runs for six-reference object binding, identity and garment transfer, and product preservation across a scene change.
- Useful editing scope
Relighting, cross-ratio outpainting, and in-image text replacement appear in the tested scenario set, covering several common revision tasks.
- 4K output
The listed maximum resolution is 4K, supporting image work that needs a larger final output.
Ограничения
- Consistency is not yet measured
Each scenario has only a selected run 1 marked by the rule "auto:only-run," so the evidence does not establish repeatability across multiple attempts.
- No comparative quality result
The evidence records scenarios and selected-run times, but provides no quality scores or direct comparisons with sibling models.
- Reference-heavy edits can take longer
The six-reference object binding and edit selection took 89.7s, longer than the other supplied selections.
- Text accuracy still needs review
Typography and in-image text replacement were exercised, but the evidence does not establish reliable spelling across languages or layouts.
Когда стоит выбрать Seedream 5 Lite
- You need to coordinate several source images in one edit.
- You want to preserve a product while changing its surrounding scene.
- You are transferring identity or garments between references.
- You need relighting, outpainting, or in-image text replacement.
- You want output up to 4K.
Когда стоит выбрать другую модель
- You need demonstrated repeatability across several runs.
- You need evidence of an advantage over another Banana Pie image model.
- Your workflow depends on verified spelling or typography accuracy.
- You need predictable turnaround for reference-heavy edits.
В зависимости от задачи лучше могут подойти: Nano Banana 2 · Seedream 5 Pro
Результаты реальных тестов
Те же фиксированные промпты, которые мы используем для каждой модели: по одному опубликованному результату с доступными для просмотра промптами и настройками.
Промпт
A premium product photograph of exactly one transparent rectangular perfume bottle, half filled with amber liquid, standing upright on wet black stone. Light comes from the upper left, creating coherent refraction, a contact shadow, and one subtle reflection. No text, logo, plants, or extra objects.
Настройки
- Resolution: 2K
- Aspect Ratio: 1:1
Промпт
Design a clean cream-colored vertical event poster. Show exactly four centered text lines and no other text: "MOONLIGHT MARKET", "月光市集", "18 OCT", and "RIVER HALL". Render "月光市集" in red and all other lines in black.
Настройки
- Resolution: 2K
- Aspect Ratio: 2:3
Промпт
A photorealistic close-up of an adult potter shaping a clay bowl on a spinning wheel. Both hands are fully visible, each with five natural fingers touching the clay. No other people or hands. Soft window light.
Настройки
- Resolution: 2K
- Aspect Ratio: 3:4
Промпт
On a matte gray table, exactly three red wooden cubes form a row on the left, exactly two blue glass spheres sit on the right, and one yellow ceramic mug stands centered behind them. The mug handle points right. No other objects or text.
Настройки
- Resolution: 2K
- Aspect Ratio: 4:3
Промпт
Create a vertical social ad for a fictional running shoe named AERO. Keep the top 15 percent empty. Directly below it, place the headline "RUN LIGHT" in the upper-left. Show exactly one silver shoe in the lower-right, an orange trail curving from the bottom-left, and a round badge reading "42 KM". No other shoes or text.
Настройки
- Resolution: 2K
- Aspect Ratio: 9:16
Промпт
Create a clean horizontal pricing comparison graphic titled "CHOOSE YOUR PLAN". Use exactly three equal columns labeled "STARTER", "PRO", and "TEAM". Under each column, show exactly three aligned rows labeled "PROJECTS", "STORAGE", and "SUPPORT", followed by one blue button labeled "SELECT". White background, dark navy text, no additional columns or text.
Настройки
- Resolution: 2K
- Aspect Ratio: 16:9
Референсные изображения
Промпт
Use Reference 1 as the base tabletop scene. Replace the red mug with the exact dark navy-blue mug from Reference 2 in the same position and orientation. Replace the folded white towel with the exact green-and-white striped napkin from Reference 3 in the same folded area. Replace the green pear with exactly one orange from Reference 4 in the same position. Replace the blue notebook with the exact mustard-yellow hardcover notebook from Reference 5 in the same position. Add exactly the two lemons from Reference 6 to the right of the mug. Preserve the wooden table, camera, framing, wood grain, lighting, shadows, and all other spatial relationships from Reference 1. Do not copy the white product backgrounds from References 2–6, duplicate any asset, or add other objects.
Настройки
- Resolution: 2K
- Aspect Ratio: 4:3
Референсные изображения
Промпт
Replace only the sign text with exactly "NIGHT OWL". Preserve the original font style, spacing, perspective, sign material, lighting, and everything else.
Настройки
- Resolution: 2K
- Aspect Ratio: 3:2
Референсные изображения
Промпт
Use the portrait in Reference 1 for the person and the isolated jacket in Reference 2 for the garment. Dress the person from Reference 1 in the exact jacket shown in Reference 2. Preserve the person's identity, face, expression, skin, hair, hands, pose, body proportions, background, framing, and lighting from Reference 1. Preserve the jacket's material, color, collar, buttons, pockets, and sleeve patch from Reference 2. Do not copy the ghost mannequin or white product background.
Настройки
- Resolution: 2K
- Aspect Ratio: 3:4
Референсные изображения
Промпт
Change the lighting to warm golden-hour sunlight entering from the left window. Do not move, add, remove, or redesign any object. Update highlights and shadows coherently.
Настройки
- Resolution: 2K
- Aspect Ratio: 16:9
Референсные изображения
Промпт
Expand the canvas to a 16:9 landscape by naturally continuing the beach, ocean, and sky on both sides. Keep the complete original image centered without cropping, stretching, letterboxing, or modifying it.
Настройки
- Resolution: 2K
- Aspect Ratio: 16:9
Референсные изображения
Промпт
Place the sneaker on a wet outdoor basketball court at dusk. Preserve the exact sneaker shape, black geometric side mark, materials, stitching, sole geometry, and camera angle. Add physically coherent contact, reflections, and dusk lighting.
Настройки
- Resolution: 2K
- Aspect Ratio: 1:1
Кредиты и цены
Generations start at 20 credits. That is the minimum listed cost for using Seedream 5 Lite; the supplied facts do not specify how settings, resolution, or reference count may affect the final credit charge.
Частые вопросы
Is Seedream 5 Lite available now?
Yes. Seedream 5 Lite is currently available on Banana Pie, with generations starting at 20 credits.
How many reference images can I use?
It supports up to 8 reference images, making it suitable for workflows that need to coordinate products, people, garments, or other source assets.
What is the maximum output resolution?
The listed maximum resolution is 4K.
Can it generate and replace text inside images?
The evidence includes selected runs for in-image text replacement and multilingual typography. These selections show that the workflows were exercised, but the available evidence does not include quality scores or repeat-run consistency data.
How long does a generation take?
Selected-run times in the supplied evidence range from 30.0s for counting and attribute binding to 89.7s for a six-reference object binding and edit. Actual time can depend on the task.
Похожие сравнения
Методика тестирования и раскрытие информации
Conclusions are based on a fixed prompt suite covering the supplied scenarios and their published selected runs; because each entry is an auto-selected only run, the draft treats coverage and timing as evidence without inferring comparative quality or repeatability.
Banana Pie предоставляет платный доступ к этой и другим моделям в одной Студии. Наши выводы основаны на тестах в той же среде, которой пользуются наши пользователи.