Image generations

OpenAI-compatible image generation endpoint.

POST https://api.ecohash.com/v1/images/generations

Headers

Header	Value
Authorization	Bearer eco_YOUR_KEY
Content-Type	application/json

Request body

Field	Type	Required	Notes
model	string	yes	Image model ID (e.g. flux2-klein, qwen-image, z-image-turbo)
prompt	string	yes	Text prompt — up to ~1000 characters
n	integer	no	Number of images. Currently only 1 is supported
size	string	no	512x512, 768x768, 1024x1024, 1024x768, 768x1024. Default 1024x1024
response_format	string	no	"b64_json" (default) or "url"
steps	integer	no	Denoising steps. Omit to use the model's default (turbo models use a low count automatically)
seed	integer	no	Deterministic generation (non-OpenAI extension)

Example request

curl https://api.ecohash.com/v1/images/generations \
  -H "Authorization: Bearer eco_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "flux2-klein",
    "prompt": "a watercolor painting of a fox in a misty forest",
    "size": "1024x1024",
    "response_format": "b64_json"
  }'

Response body

{
  "created": 1776391234,
  "data": [{
    "b64_json": "iVBORw0KGgo...",
    "revised_prompt": null
  }]
}

With response_format: "url", the response has url: "https://..." instead of b64_json — URL is valid for 24 hours.

Image edits

POST https://api.ecohash.com/v1/images/edits

Edit an existing image with a prompt. Send as multipart/form-data with an edit-capable model (for example flux2-klein, qwen-image).

Field	Type	Required	Notes
model	text	yes	An edit-capable image model
image	file	yes	The source image
prompt	text	yes	The change to apply
size	text	no	Output size

curl https://api.ecohash.com/v1/images/edits \
  -H "Authorization: Bearer eco_YOUR_KEY" \
  -F model=flux2-klein \
  -F image=@input.png \
  -F prompt="make the sky a dramatic sunset" \
  -F size=1024x1024

The response shape matches /v1/images/generations. A text-to-image-only model returns an error here.

Errors

See Errors. Common: 401, 402, 404, 429.