A whole series in one request
Set the “Images” counter from 1 to 15 and describe the series in a single prompt. ByteDance built this mode for sets of related images, such as comic storyboards or brand visuals.
Image model · ByteDance Seed
Seedream v5.0 Lite Sequential — A consistent series of images from a single prompt.
Seedream v5.0 Lite Sequential is the series mode of Seedream 5.0 Lite, the image model ByteDance’s Seed team introduced in February 2026. On Deepnia you ask for up to 15 images at once, designed to share the same characters, style and aspect ratio: a carousel, a storyboard, a comic strip or a sticker pack, from text or up to 10 reference images.
- Series of 1 to 15 images
- Up to 10 references
- 7 ratios in 2K
- One ratio for the whole series
- Generation and editing
- 3,000-character prompt
The image tool opens with Seedream v5.0 Lite Sequential already selected. The exact cost shows before every creation; you sign up when you generate.
- A prompt describing each image
- Up to 10 references (optional)
- The number of images, 1 to 15
- A consistent series in one aspect ratio
At a glance
- Maker
- ByteDance Seed
- Images per request
- 1 to 15 (2 by default)
- References
- Optional, up to 10 images
- Aspect ratios
- 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3 (1:1 by default)
- Size
- 2K, e.g. 2048×2048 in 1:1
- Prompt
- 3 to 3,000 characters
What Seedream v5.0 Lite Sequential does best
The same characters from image to image
The series mode is built to keep characters, settings and art direction from one image to the next. ByteDance shows, for example, the same characters in the morning, at noon and at night.
A brand identity from a logo
Add a logo, a product or a mascot (up to 10 images): ByteDance shows a complete visual system, from packaging to cards, created from a single logo.
The same aspect ratio across the series
Every image shares the ratio you pick among the 7 on offer, at 2K size: handy for a 3:4 carousel or a 16:9 storyboard.
Built on Seedream 5.0 Lite
The series mode runs on Seedream 5.0 Lite, which ByteDance presents as a model that grasps the intent behind instructions, borrows a style from one image and handles scenes with several characters.
How to use Seedream v5.0 Lite Sequential on Deepnia
Open the image tool
The “Try Seedream v5.0 Lite Sequential” button opens Deepnia’s image generator with the model already selected.
Set the number of images
With the “Images” counter, choose 1 to 15 images (2 by default).
Add references if needed
A logo, a product or a character: up to 10 images with “Add an image” under “References”. Without an image, the series starts from text.
Describe the series and run it
State the number of images and describe each one, pick the aspect ratio, check the cost shown, then click “Generate”, or “Edit” when you use references.
Seedream v5.0 Lite Sequential on Deepnia: what you can set
- Images per request
- 1 to 15, 2 by default, with the “Images” counter, which sets the maximum for the series
- Inputs
- Text only, or text and reference images (“Edit mode”)
- References
- Up to 10 images; JPG, PNG or WebP, up to 10 MB each
- Aspect ratios
- 1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3 (1:1 by default), the same for the whole series
- Requested size per ratio
- 2048×2048 (1:1), 2560×1440 (16:9), 1440×2560 (9:16), 2560×1920 (4:3), 1920×2560 (3:4), 2560×1707 (3:2), 1707×2560 (2:3)
- Prompt
- Required, 3 to 3,000 characters
- References and created images
- 15 in total per request, according to ByteDance
Prompt ideas to try
Copy a prompt, open the tool and adapt it to your project.
Carousel for a fashion brand
- 5 images
- 3:4
- 1 reference image: the fashion model
Create a series of 5 consistent images for a fashion carousel: the woman from the reference image wears 5 different outfits (long blue dress, orange skirt and top set, green jumpsuit, loose yellow shirt, red and gold evening gown). Same beige studio, same soft lighting, same full-length framing, in this order.
Storyboard for a coffee shop ad
- 6 images
- 16:9
- Text only
Generate a series of 6 storyboard frames, realistic illustration style, for a neighborhood coffee shop ad: 1) the barista opens the shutter at dawn; 2) he grinds the coffee; 3) milk foams in the pitcher; 4) the room fills up mid-morning; 5) two friends clink their cups; 6) final shot of the storefront lit up at night. Same characters and same coffee shop in every frame.
Brand kit for an Accra bakery
- 6 images
- 1:1
- 1 reference image: the logo
Using the logo from the reference image, create a series of 6 consistent visuals for a bakery in Accra: kraft paper bag, cake box, apron, shop sign, loyalty card, coffee cup. Logo colors, warm modern style, plain light background.
Comic strip for a classroom
- 4 images
- 3:2
- Text only
Create a 4-panel comic for children about handwashing: 1) Leo comes back from recess with dirty hands; 2) he washes his hands with soap at the school sink; 3) he scrubs between his fingers while counting to 20; 4) he eats his snack with a smile. Same character, colorful illustration style, no speech bubbles or text.
Travel diary in four moments
- 4 images
- 9:16
- Text only
A series of 4 vertical images with the same traveler (short curly hair, denim jacket, red backpack) during one day in Lisbon: 1) coffee and a pastry on a terrace in the morning; 2) a ride on a yellow tram; 3) a viewpoint at sunset; 4) a candlelit dinner in a narrow street. Same realistic photo style, natural light.
Sticker pack for a mascot
- 8 images
- 1:1
- 1 reference image: the mascot
Using the mascot from the reference image, create 8 consistent stickers on a white background: waving, laughing, crying with joy, winking, thumbs up, surprised, sleeping, celebrating with confetti. Same style, same line work, thick white outline.
Tips for better results
- State the number of images in the prompt and describe each one in order (1, 2, 3…): according to ByteDance, the model decides how many images to return.
- Set the “Images” counter to the same number as in the prompt: it sets the maximum for the series.
- Describe once what stays the same (character, outfit, place, style, light), then what changes from one image to the next.
- To keep a character or a product identical, add a clear reference image of it rather than a description alone.
- Pick the aspect ratio for where the series will go: 3:4 or 1:1 for a carousel, 16:9 for a storyboard, 9:16 for stories.
Good to know
- References and created images count together, 15 in total per request according to ByteDance: with 10 references, plan for up to 5 images.
- For text in the images (a sign, a label, a title), put it in quotation marks, keep it short and check every word.
- For a standalone image or a targeted edit, switch to Seedream v5.0 Lite: the same model, one image per creation.
- Use logos, product photos and portraits you have the rights to, with the consent of the people pictured.
Seedream v5.0 Lite Sequential or another model?
Deepnia’s other image models, to pick the right one for your project.
| Model | Resolution | Aspect ratios | References | Best for |
|---|---|---|---|---|
| Seedream v5.0 Lite SequentialThis model | Automatic | 7 ratios | Up to 10 | A consistent series of images from a single prompt. |
| Seedream v5.0 Lite | Automatic | 7 ratios | Up to 10 | The same model for standalone images, created or edited one at a time. |
| GPT Image 2 | 1K–4K | 8 ratios | Up to 16 | A text-rich visual (infographic, poster), up to 4K. |
| Nano Banana Pro | 1K–4K | 6 ratios | Up to 8 | High-fidelity editing with several references, up to 4K. |
| Nano Banana 2 Lite | 1K | 14 ratios | Up to 10 | Deepnia’s default image model, for a single image or an edit with references. |
Frequently asked questions
How many images can I generate at once?
1 to 15 images per request on Deepnia, 2 by default, set with the “Images” counter.
How do I get the number of images I want?
State that number in the prompt, describe each image in order (1, 2, 3…) and set the “Images” counter to the same number: in series mode, the model decides how many images to return, up to the number you chose. With references, count 15 images in total, references included.
Are the images in a series consistent?
That is the point of this mode: ByteDance presents it for sets of related images, such as comic storyboards or brand visuals, and shows the same characters from one image to the next. To strengthen consistency, add a reference image of the character or product and describe once what stays the same.
Can I start from a logo or a product photo?
Yes: add up to 10 reference images (JPG, PNG or WebP, up to 10 MB). ByteDance shows a complete visual identity created from a single logo.
Which aspect ratios are available?
1:1, 16:9, 9:16, 4:3, 3:4, 3:2, 2:3, at 2K size (for example 2048×2048 in 1:1); the whole series uses the same ratio.
What is the difference with Seedream v5.0 Lite?
It is the same ByteDance model: Seedream v5.0 Lite creates standalone images, while the Sequential version creates a linked series of 1 to 15 images in a single request.
Which aspect ratio should I pick for a carousel, a storyboard or stories?
3:4 or 1:1 for a carousel, 16:9 for a storyboard, 9:16 for stories: the whole series is created in that ratio, at 2K.
What is it good for?
Social media carousels, ad storyboards, comic strips, sticker packs, travel diaries and brand identity sets.
Ready to try Seedream v5.0 Lite Sequential?
The image tool opens with Seedream v5.0 Lite Sequential already selected. The exact cost shows before every creation; you sign up when you generate.