Seedream 4.5 Text to Image API

bytedance/seedream/v4.5/text-to-image

2K/4K · 8 ratios · Custom size

Seedream 4.5 Text to Image transforms text prompts into 2K and 4K high-fidelity visuals, supporting crisp bilingual typography, refined portrait lighting, and batch exploration of 1–15 images. It follows spatial semantics and prompt instructions closely while maintaining balanced compositions across diverse aspect ratios for ready-to-use production assets.

Get API Key

Input

PromptRequired0/3000

SizeOptional

NOptional

Enable Safety CheckerOptional

OutputIdle

Generated images will appear here

Estimated cost: —

Continue using

Seedream 4.5 Image Edit Seedream 5.0 Pro Text to Image Seedance 2.5 Text to Video

Examples

A detailed isometric cutaway illustration of a small polar research station standing on steel stilts on an Antarctic ice shelf, shown like a page from a science museum guidebook. The roof and front wall are removed so six rooms are visible, each with a small clean sans-serif label plate printed exactly: "LAB" (a scientist examining an ice core under a lamp), "GALLEY" (two people cooking soup at a steaming stove), "BUNKS" (stacked beds with colorful quilts), "RADIO" (an operator with headphones in front of dials), "GREENHOUSE" (lettuce and tomatoes under pink grow lights) and "GARAGE" (an orange snowcat being repaired). A banner across the top reads exactly "AURORA RIDGE STATION", and a small thermometer sign beside the entrance reads "-38°C". Tiny figures, pipes, ladders and cables connect the rooms logically. Outside: blowing snow, a weather mast, a pale green aurora in the dark sky. Crisp linework, soft flat colors, consistent isometric perspective, every label spelled correctly and legible, no extra text, no watermark.

A vertical documentary portrait photograph of an elderly South Indian fisherwoman mending a fishing net under the palm-thatched roof of a wooden pier in Kerala during a monsoon evening. She sits cross-legged, wearing a faded turquoise cotton sari with a thin gold border and a small silver nose stud; her deeply lined face, silver hair tied back and weathered hands pulling green nylon mesh are in sharp focus. Blue-hour light from the backwaters mixes with the warm glow of a single hanging kerosene lantern beside her. Heavy rain falls in streaks beyond the roof edge, drops glint on the net, wet planks reflect the lantern, and a few moored wooden boats fade into the misty background. Natural skin texture with pores and fine wrinkles, calm dignified expression looking at her work, 85mm lens, shallow depth of field, believable hands with five fingers, no text, no watermark.

An ultra-wide cinematic panorama from inside a gigantic rotating cylindrical space habitat. The camera stands on a gravel path in the foreground among ripening wheat and small orchards; the farmland, rivers, winding roads and small white villages continuously curve upward on both sides and arc overhead, so the opposite side of the cylinder is visible high in the sky, upside down, with tiny fields and lakes. A long glowing sun-line runs along the central axis, casting soft morning light and thin wispy clouds drifting in the middle of the cylinder. At the far end, a huge circular end-cap window shows black space, stars and a slice of a blue planet. A cyclist rides along the path for scale. Coherent curved perspective, consistent lighting, atmospheric haze with distance, grounded realistic sci-fi concept art with photographic detail, no text, no logos, no watermark.

REST API

Quick Start

This flow submits one valid Seedream 4.5 Text to Image request, captures its task_id, and reads image URLs after the asynchronous task ends.

Connect to the Vidgo API

Create an API key, keep it on the server, and send it as Authorization: Bearer VIDGO_API_KEY.

Endpoint: POSThttps://api.vidgo.ai/api/generate/submit
Authentication: Authorization: Bearer VIDGO_API_KEY

Submit one image task

Send the current model ID, prompt to the unified submit endpoint. The example keeps only first-call fields.

Submit request

curl --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data '{
  "model": "bytedance/seedream/v4.5/text-to-image",
  "input": {
    "prompt": "A detailed isometric cutaway illustration of a small polar research station standing on steel stilts on an Antarctic ice shelf, shown like a page from a science museum guidebook. The roof and front wall are removed so six rooms are visible, each with a small clean sans-serif label plate printed exactly: \"LAB\" (a scientist examining an ice core under a lamp), \"GALLEY\" (two people cooking soup at a steaming stove), \"BUNKS\" (stacked beds with colorful quilts), \"RADIO\" (an operator with headphones in front of dials), \"GREENHOUSE\" (lettuce and tomatoes under pink grow lights) and \"GARAGE\" (an orange snowcat being repaired). A banner across the top reads exactly \"AURORA RIDGE STATION\", and a small thermometer sign beside the entrance reads \"-38°C\". Tiny figures, pipes, ladders and cables connect the rooms logically. Outside: blowing snow, a weather mast, a pale green aurora in the dark sky. Crisp linework, soft flat colors, consistent isometric perspective, every label spelled correctly and legible, no extra text, no watermark.",
    "size": "4:3",
    "n": 1,
    "enable_safety_checker": true
  }
}'

Wait for the image result

Read data.task_id and poll only while status is not_started or running. Stop on both finished and failed.

Track status

GET https://api.vidgo.ai/api/generate/status/{task_id}

Poll every 2–5 seconds and back off on 429/5xx. Or provide top-level callback_url to receive a terminal-state callback.

not_startedrunningfinishedfailed

Submit response

{
  "code": 200,
  "data": {
    "task_id": "PGQCZ3X31DC6OEXQ",
    "status": "running",
    "created_time": "2026-09-25T16:14:03"
  }
}

Finished response

{
  "code": 200,
  "data": {
    "task_id": "PGQCZ3X31DC6OEXQ",
    "status": "finished",
    "files": [
      {
        "file_url": "https://cdn.vidgo.ai/apis/models/bytedance/seedream/v4.5/text-to-image/v1/01/output.jpg",
        "file_type": "image"
      }
    ],
    "created_time": "2026-09-25T16:14:03",
    "error_message": null,
    "progress": 100
  }
}

Complete runnable polling example

Expand for the same payload end to end, including business-code, missing-task-ID, failed-state, unexpected-state, and network-error handling.

Complete runnable polling example

set -euo pipefail
: "${VIDGO_API_KEY:?Set VIDGO_API_KEY in your environment}"

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "bytedance/seedream/v4.5/text-to-image",
  "input": {
    "prompt": "A detailed isometric cutaway illustration of a small polar research station standing on steel stilts on an Antarctic ice shelf, shown like a page from a science museum guidebook. The roof and front wall are removed so six rooms are visible, each with a small clean sans-serif label plate printed exactly: \"LAB\" (a scientist examining an ice core under a lamp), \"GALLEY\" (two people cooking soup at a steaming stove), \"BUNKS\" (stacked beds with colorful quilts), \"RADIO\" (an operator with headphones in front of dials), \"GREENHOUSE\" (lettuce and tomatoes under pink grow lights) and \"GARAGE\" (an orange snowcat being repaired). A banner across the top reads exactly \"AURORA RIDGE STATION\", and a small thermometer sign beside the entrance reads \"-38°C\". Tiny figures, pipes, ladders and cables connect the rooms logically. Outside: blowing snow, a weather mast, a pale green aurora in the dark sky. Crisp linework, soft flat colors, consistent isometric perspective, every label spelled correctly and legible, no extra text, no watermark.",
    "size": "4:3",
    "n": 1,
    "enable_safety_checker": true
  }
}
JSON
)

SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

TASK_ID=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.data.task_id // .task_id // empty')
if [ -z "$TASK_ID" ]; then
  printf 'Submit response did not include task_id:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi

while true; do
  STATUS_RESPONSE=$(curl --silent --show-error --fail-with-body \
    --url "https://api.vidgo.ai/api/generate/status/$TASK_ID" \
    --header "Authorization: Bearer $VIDGO_API_KEY")
  STATUS=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.data.status // .status // empty')

  case "$STATUS" in
    finished)
      printf '%s' "$STATUS_RESPONSE" | jq -r '(.data.files // .files // [])[]?.file_url'
      break
      ;;
    failed)
      printf '%s' "$STATUS_RESPONSE" | jq -r '.data.error_message // .error_message // "Generation failed"' >&2
      exit 1
      ;;
    not_started|running)
      sleep 2
      ;;
    *)
      printf 'Unexpected task status: %s
' "$STATUS" >&2
      exit 1
      ;;
  esac
done

Request Parameters

This table lists only public fields inside input. See Quick Start for top-level model and optional callback_url usage.

Field	Type	Required	Default	Description
input.prompt	string	Yes	–	Describe the image, art style, and composition in 1–3,000 characters.
input.size	string \| object	No	1:1	Output size: 2K, 4K, or an aspect ratio. For a custom size, enter WIDTHxHEIGHT (for example 1920x4096) or {"width":2304,"height":3072} with positive integers. Defaults to 1:1.
input.n	integer	No	1	Number of images to generate: integer 1–15. Defaults to 1.
input.enable_safety_checker	boolean	No	true	Set true to enable safety checking or false to turn it off. Defaults to true.

Response Fields

The submit response identifies the task. The status response supplies image files or failure details at a terminal state.

Field	Type	Description
code	integer	Business response code; 0 or 200 indicates success.
message	string	Optional human-readable response message.
data.task_id	string	Unique task ID used for status queries after submission.
data.status	string	not_started, running, finished, or failed.
data.created_time	string	Task creation timestamp.
data.progress	integer	Progress from 0–100 in status responses.
data.files[]	array	Generated files in returned order when finished.
data.files[].file_url	string	Direct generated-image URL.
data.files[].file_type	string	Image results are identified as image.
data.files[].watermark_url	string \| null	Optional watermarked-file URL when present.
data.error_message	string \| null	Readable failure reason for failed tasks.

Task Lifecycle

Only the first two states may be polled again. finished and failed are terminal.

not_started
Accepted and queued; wait before checking again.
running
Image generation is in progress; continue polling.
finished
Successful terminal state; validate files and read image URLs.
failed
Failure terminal state; read error_message and stop.

Polling and Errors

AuthenticationKeep a valid VIDGO_API_KEY on the server and send requests with Bearer authentication.
Validation400 indicates an invalid model ID, field, enum, length, or image count.
Polling intervalStart at 2–5 seconds, use exponential backoff for 429/5xx, and enforce an overall timeout.
Terminal statesRead image URLs from files after finished, or error_message after failed. Both states end polling.
Callback optionProvide a public top-level callback_url to receive a flat task object after finished or failed.
Network errorsDistinguish request timeout, network failure, and task failure. Retry the same query only when safe.

Model Specifications

Specification	Value	Details
Model	bytedance/seedream/v4.5/text-to-image	Text-to-image generation.
size	1:1 / 4:3 / 3:4 / 16:9 / 9:16 / 3:2 / 2:3 / 21:9 / 2K / 4K	Default: 1:1. Custom WIDTHxHEIGHT or {"width","height"} is also accepted.
n	1–15	Default: 1.

Seedream 4.5 Text to Image API documentation

Seedream 4.5 Text to Image

Seedream 4.5 Text to Image is a high-fidelity text-to-image generation model developed by the ByteDance Seed team. Sharing a unified architecture with its companion image editing endpoint and scaled up from Seedream 4.0, it delivers substantial upgrades across spatial reasoning, 3D depth, and complex prompt execution. The model natively generates 2K and 4K images across eight aspect ratios and custom dimensions with zero high-resolution surcharge, excelling at dense bilingual typography and realistic facial details even on small-scale subjects. Generating 30%–40% faster and supporting 1 to 15 batch variations, it powers production pipelines in commercial advertising, e-commerce, and creative design.

Why Choose Seedream 4.5 Text to Image API?

Breakthrough Bilingual and Dense Small-Text TypographyRenders legible, sharp Chinese and English text, dense product packaging copy, and brand logos directly within visuals, overcoming typical AI character distortion for production-ready designs.
Native 4K Output with Zero Surcharge and Small-Face ClarityProduces native 4K ultra-high-definition visuals without resolution markups, capturing realistic skin micro-textures and cinematic lighting while preserving natural clarity even on small-scale faces.
3D Spatial Reasoning and Compositional PerspectiveLeverages upgraded spatial comprehension to maintain accurate physical proportions, depth of field, and object occlusion, ensuring balanced, coherent perspective across multi-subject scenes.
1–15 Batch Output with 30%–40% Generation SpeedupGenerates up to 15 distinct framing and angle variations in a single run with 30%–40% faster generation speeds than version 4.0, drastically reducing creative turnaround time at 5 credits per image.

Parameters

Parameter	Requirement	Description
prompt	Required	Describe the image, art style, and composition in 1–3,000 characters.
size	Optional	Output size: 2K, 4K, or an aspect ratio. For a custom size, enter WIDTHxHEIGHT (for example 1920x4096) or {"width":2304,"height":3072} with positive integers. Defaults to 1:1. Default`1:12K4K4:33:416:99:163:22:321:9`
n	Optional	Number of images to generate: integer 1–15. Defaults to 1. Default`123456789101112131415`
enable_safety_checker	Optional	Set true to enable safety checking or false to turn it off. Defaults to true. Default`true`

How to Use

Craft Your Prompt and Text ContentDescribe the scene, subject, aesthetic style, lighting, and composition in detail. Enclose any required text in quotation marks to guide placement and spelling.
Select Dimensions and Output CountChoose a preset aspect ratio such as 1:1, 16:9, or 9:16, specify 2K or 4K resolution, or provide custom dimensions, and set the image count n between 1 and 15.
Submit Task and Retrieve ResultsSend your request to the unified endpoint to receive a task_id, then poll status or configure a callback_url to access the generated image URLs.

Pricing

Each generation costs 5 credits ($0.025). n is the number of generated images; the total is 5 × n credits.

Usage	Rate	Details
Standard generation	5 credits / generation · $0.025 / generation	5 × n credits · $0.025 × n

Best Use Cases

Commercial Ads and Marketing PostersGenerate complete promotional creatives featuring legible bilingual headlines, promotional tags, and polished product arrangements.
E-Commerce Product Scenes and Lifestyle VisualsProduce photorealistic contextual backgrounds and lifestyle mockups tailored to catalog items with physically accurate ambient lighting.
Concept Art and Editorial MediaExplore diverse creative variations across 1–15 images for film pre-production, book covers, social media graphics, and digital storytelling.

Pro Tips

Typography prompting: Enclose exact slogans or titles in quotation marks (such as "NEW ARRIVAL" at the top) and explicitly state font weight, placement, color, and language.
Activate 3D spatial depth: For complex multi-subject scenes, specify foreground and background relationships (e.g. "foreground right", "deep center") to direct 3D depth and proportions.
Staged creation workflow: Explore creative directions at 2K or with n set to 4 for rapid iteration, then produce final deliverables at native 4K with no resolution surcharge.

Notes

Text-only input: This endpoint accepts text prompts only. For reference-guided editing or multi-image composition, use the Seedream 4.5 Image Edit endpoint.
Prompt length limit: Prompts must contain between 1 and 3,000 characters after trimming whitespace.
Billing and refunds: Tasks reserve 5 credits ($0.025) per requested image. Unused credits are automatically refunded for failed tasks or when fewer than n images are returned.

Related Models

Grok Imagine Image 2.0 Text to Image preview

Grok Imagine Image 2.0 Text to ImageGenerate crisp, high-fidelity images up to 2K from text prompts with Grok Imagine Image 2.0. Features designer-grade typography, precise prompt adherence, and versatile aspect ratios.

Grok Imagine Image 2.0 EditGenerate crisp, high-fidelity images up to 2K from text prompts with Grok Imagine Image 2.0. Features designer-grade typography, precise prompt adherence, and versatile aspect ratios.

Qwen Image 3.0 Text-to-ImageAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Image EditAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Pro Text-to-Image preview

Qwen Image 3.0 Pro Text-to-ImageAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Qwen Image 3.0 Pro Image EditAlibaba Qwen Image 3.0 Standard and Pro generation and editing with eight aspect ratios, 1K or 2K output, and ordered one-to-three-image editing.

Seedream 5.0 Pro Text-to-ImageThe ByteDance Seedream 5.0 Pro API generates 1K or 2K JPEG/PNG images from prompts and edits one to ten ordered HTTP(S) image inputs.

Seedream 5.0 Pro Image EditThe ByteDance Seedream 5.0 Pro API generates 1K or 2K JPEG/PNG images from prompts and edits one to ten ordered HTTP(S) image inputs.

Nano Banana 2 Lite Text to Image preview

Nano Banana 2 Lite Text to ImageGenerate crisp 1K images in ~4 seconds from text prompts with Nano Banana 2 Lite. Optimized for high-volume rapid iteration, in-image text in 25+ languages, and 10 aspect ratios at $0.025 per generation.

Nano Banana 2 Lite EditGenerate crisp 1K images in ~4 seconds from text prompts with Nano Banana 2 Lite. Optimized for high-volume rapid iteration, in-image text in 25+ languages, and 10 aspect ratios at $0.025 per generation.

Grok Imagine Image Quality Text to Image preview

Grok Imagine Image Quality Text to ImageGenerate high-fidelity images up to 2K resolution from text prompts with Grok Imagine Image Quality. Features photorealistic detail, sharp text rendering, and 5 aspect ratios.

Grok Imagine Image Quality EditGenerate high-fidelity images up to 2K resolution from text prompts with Grok Imagine Image Quality. Features photorealistic detail, sharp text rendering, and 5 aspect ratios.

GPT Image 2 Text-to-ImageOpenAI image generation and editing with quality tiers, flexible size controls, and single-image outputs up to 4K.

GPT Image 2 Image EditOpenAI image generation and editing with quality tiers, flexible size controls, and single-image outputs up to 4K.

Wan 2.7 Image Standard Text to Image preview

Wan 2.7 Image Standard Text to ImageCreate and edit images from prompts or reference photos with Wan 2.7 Image Standard and Pro models.

Wan 2.7 Image Standard EditCreate and edit images from prompts or reference photos with Wan 2.7 Image Standard and Pro models.

Seedream 4.5 Text to Image API frequently asked questions

What is the Seedream 4.5 Text to Image API?

Seedream 4.5 Text to Image is a ByteDance Seed model for high-quality text-to-image generation. It generates 2K and 4K ultra-high-definition images from text prompts with crisp bilingual typography, refined portrait lighting, and batch creation of 1–15 images. Built on an advanced global scaling multimodal foundation, it preserves spatial semantics and prompt adherence while maintaining balanced compositions across diverse aspect ratios. You can call it programmatically or try it from the playground above.

Can Seedream 4.5 Text to Image generate multiple images at once?

Yes, you can generate 1 to 15 independent images in a single call. By configuring the n parameter from 1 to 15, you receive multiple composition and framing variations for the same prompt, billed only for the images successfully produced.

Does Seedream 4.5 Text to Image support rendering Chinese and English text?

Yes. The model features a built-in typography engine that accurately renders legible Chinese and English characters, headlines, and slogans directly within generated graphics when enclosed in quotation marks.

What resolutions and aspect ratios does Seedream 4.5 Text to Image support?

It supports 2K and 4K resolution tiers along with eight standard aspect ratio presets including 1:1, 4:3, 3:4, 16:9, 9:16, 3:2, 2:3, and 21:9, as well as custom positive integer dimensions such as 1920x4096 or an object.

How does Seedream 4.5 Text to Image handle content safety?

An automated safety checker is enabled by default through enable_safety_checker set to true. It screens requests to ensure safe visual content, and credits are automatically refunded if a task produces no output due to moderation.

What is the typical generation latency for Seedream 4.5 Text to Image?

Benefiting from architectural scaling, Seedream 4.5 generates 30%–40% faster than version 4.0. Under standard cluster loads, end-to-end generation latency for a 2K or 4K image has a median of 5 to 15 seconds, and production applications should poll every 2 to 5 seconds or configure a callback_url.

How are credits handled if a Seedream 4.5 Text to Image task fails?

Vidgo uses an automated refund guarantee. Tasks pre-charge 5 credits per image upon submission; if a task fails due to validation errors or timeouts, all credits are returned, and partial output is refunded proportionally.