Images API

Painting models on Omniall AI mainly use these entry points:

Style	Path	Typical models
OpenAI Images (recommended)	`POST /v1/images/generations` · `/v1/images/edits`	GPT-Image, Flux, Grok Imagine, Seedream / Jimeng, many Gemini image models
Gemini Chat	`POST /v1/chat/completions`	`gemini-nano-banana-2.1` (image / PDF / short-video refs), Flash/Pro image models
Gemini native	`POST /v1beta/models/{model}:generateContent`	Same; include `IMAGE` in `responseModalities`; Banana 2.1 accepts PDF / MP4 `inlineData`
Midjourney	`/mj/submit/*` + `GET /mj/task/{id}/fetch`	Midjourney via MJ proxy

Host: https://api.omniall.ai · Auth Authorization: Bearer sk-... · models: Model Square. Banana 2.1 PDF / video examples: Gemini native format and Chat image generation.

Quick example

curl https://api.omniall.ai/v1/images/generations \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-1",
    "prompt": "A calm lake at dawn",
    "size": "1024x1024"
  }'

Upload reference files first with File upload when you need a public URL. Field tables and playgrounds: API Reference → Images.