Image generation API

Seedream

ByteDance's Seedream v4/v4.5/v5 family — high-quality image generation with sequential and edit modes for iterative refinement.

Provider: ByteDance16 models

Available models

Model key	Display name	Capability	Atlas Cloud slug
`seedream_5_pro_t2i`	Seedream 5 Pro T2I	text-to-image	bytedance/seedream-v5.0-pro/text-to-image
`seedream_5_pro_edit`	Seedream 5 Pro Edit	edit	bytedance/seedream-v5.0-pro/edit
`seedream_5_pro_seq`	Seedream 5 Pro Sequential	sequential	bytedance/seedream-v5.0-pro/sequential
`seedream_5_pro_edit_seq`	Seedream 5 Pro Edit Seq	edit-sequential	bytedance/seedream-v5.0-pro/edit-sequential
`seedream_v5_lite`	Seedream 5 Lite	text-to-image	bytedance/seedream-v5.0-lite
`seedream_v5_lite_edit`	Seedream 5 Lite Edit	edit	bytedance/seedream-v5.0-lite/edit
`seedream_v5_lite_seq`	Seedream 5 Lite Sequential	sequential	bytedance/seedream-v5.0-lite/sequential
`seedream_v5_lite_edit_seq`	Seedream 5 Lite Edit Seq	edit-sequential	bytedance/seedream-v5.0-lite/edit-sequential
`seedream_v4_5`	Seedream v4.5	text-to-image	bytedance/seedream-v4.5
`seedream_v4_5_edit`	Seedream v4.5 Edit	edit	bytedance/seedream-v4.5/edit
`seedream_v4_5_seq`	Seedream v4.5 Sequential	sequential	bytedance/seedream-v4.5/sequential
`seedream_v4_5_edit_seq`	Seedream v4.5 Edit Seq	edit-sequential	bytedance/seedream-v4.5/edit-sequential
`seedream_v4`	Seedream v4	text-to-image	bytedance/seedream-v4
`seedream_v4_edit`	Seedream v4 Edit	edit	bytedance/seedream-v4/edit
`seedream_v4_seq`	Seedream v4 Sequential	sequential	bytedance/seedream-v4/sequential
`seedream_v4_edit_seq`	Seedream v4 Edit Seq	edit-sequential	bytedance/seedream-v4/edit-sequential

Endpoint

POST /api/generate

Request body

Parameter	Type	Required	Description
`modelKey`	string	✓	One of the model keys listed above
`prompt`	string	✓	Text description of the image to generate
`inputUrl`	string		Input image URL (for edit/i2i models)
`aspectRatio`	string		`"1:1"`, `"16:9"`, `"9:16"`, etc.
`imageQuality`	string		`"Low"`, `"Medium"`, `"High"`
`styleRef`	string \| string[]		Style reference image URL(s)
`charRef`	string \| string[]		Character reference image URL(s)
`brandId`	string		Brand DNA ID for brand-consistent generation

Code example

curl -X POST https://api.oyefilmy.com/api/generate \
  -H "Authorization: Bearer of_live_YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "modelKey": "seedream_5_pro_t2i",
    "prompt": "A futuristic cyberpunk cityscape at night, neon-lit skyscrapers, flying cars, rain reflections"
  }'

Response

{
  "success": true,
  "predictionId": "pred_abc123...",
  "status": "processing",
  "model": "bytedance/seedream-v5.0-pro/text-to-image",
  "creditCost": 5
}

Sequential mode allows iterative refinement — chain multiple generation steps where each uses the previous output as input. Seedream 5 Pro is the highest quality option.

← All image models

Back to image generation overview

Webhooks

Receive results automatically