Image Gen API — AI Image Generation API

A developer REST API for AI image generation — send a prompt, get back hosted image URLs. Three models, an async job flow, and a free tier with an instant API key.

Base URL: https://img-gen-api.sikasio.com

Generation is job-based: you submit a prompt and the API returns a job. Poll the job until it's done, or use the ?wait=true shortcut to get the result in a single call. Generated images are hosted on our servers and returned as stable https URLs.

Quick facts

Base URL

https://img-gen-api.sikasio.com

Endpoint

POST /v1/generate → poll GET /v1/jobs/:id

Auth

X-API-Key header — instant free key at /signup

Models

flash (1 credit) · studio · vivid (2 credits)

Output

Hosted JPG URLs, 1–4 images per request, 5 aspect ratios

Pricing

Free $0 (25 credits/mo) · Starter $12/mo (400) · Pro $39/mo (2,000)

Provider

Sikasio · support@sikasio.com

Get your free API key →

Getting your API key

New here? You can be making your first request in under a minute:

1. Create a free account at /signup — just an email and a password, and accept the terms. No card required.
2. Verify your email by entering the 6-digit code we send you.
3. Copy your API key — it's shown once, right after verification, so save it somewhere safe. You can always see its prefix or regenerate a fresh key from your account dashboard.
4. Send the key as the X-API-Key header (or Authorization: Bearer) on every /v1 request.
5. Make your first call: POST /v1/generate with a prompt — see the quick start below.

Already have an account? Sign in. Forgot your password? Reset it here.

Authentication

Every /v1 request must include your API key in the X-API-Key header:

X-API-Key: igk_live_YOUR_KEY

An Authorization: Bearer igk_live_YOUR_KEY header works as an alternative to X-API-Key if that's a better fit for your HTTP client.

Missing, wrong, or disabled keys return 401. Credits are only charged for successful generations — failed jobs and validation errors cost nothing.

CORS: enabled (Access-Control-Allow-Origin: *) — you can call the API and embed returned image URLs directly from browser apps. Note that some sandboxed preview environments enforce their own content-security-policy allowlist and may block third-party hosts regardless.

Plans & quotas

Every plan has a per-minute rate limit plus hourly, daily, and monthly credit allowances (see models & credits for what a credit buys):

Check your live usage and remaining credits any time via GET /v1/usage. Upgrade from your dashboard whenever you need more.

Models & credits

How credits work: 1 image = 1 credit on flash, 2 credits on studio/vivid; charged on success only.

Model availability: each model has a daily capacity. When a model's capacity for the day is used up it is temporarily unavailable (the others keep working) and comes back at the next daily reset — usually around 10:00 UTC (13:00 Cairo). Check before you send: GET /v1/availability (public, no key) or the available fields on GET /v1/models. A request for an unavailable model is rejected up front with 503 model_unavailable — no job is created and nothing is charged (see errors). Live view: status page.

Quick start

One-shot (synchronous)

curl -X POST "https://img-gen-api.sikasio.com/v1/generate?wait=true" \
  -H "X-API-Key: igk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "prompt": "a neon coffee cup, studio product shot",
    "model": "flash",
    "count": 2,
    "aspectRatio": "1:1"
  }'

{
  "id": "ab1ed22c-...",
  "status": "done",
  "images": [
    { "url": "https://img-gen-api.sikasio.com/files/ab1ed22c-..._1_89dc.jpg",
      "seed": 1698732404, "id": "user:...", "bytes": 575942 }
  ]
}

Use each images[].url directly in an <img> tag or download it.

Async vs. synchronous

Create a generation

Query: ?wait=true (optional) — block and return the finished job.

Body (JSON):

Returns 202 (async) or 200 (with ?wait=true, when done):

{ "id": "ab1ed22c-...", "status": "queued", "model": "flash",
  "count": 2, "createdAt": "2026-07-08T12:59:41.316Z" }

Retrieve a job

status transitions queued → running → done | failed.

{
  "id": "ab1ed22c-...",
  "status": "done",
  "model": "flash",
  "count": 2,
  "createdAt": "2026-07-08T12:59:41.316Z",
  "finishedAt": "2026-07-08T13:00:38.152Z",
  "images": [
    { "url": "https://img-gen-api.sikasio.com/files/..._1_....jpg",
      "seed": 1698732404, "id": "user:...", "bytes": 575942 }
  ]
}

On failure: { "status": "failed", "error": "…reason…" }. Unknown id → 404. Jobs are visible only to the API key that created them.

List models

{ "default": "flash",
  "models": [
    { "id": "flash",  "label": "Flash",  "credits": 1, "description": "Fast standard image generation.",
      "available": true,  "reason": null,             "availableAt": null },
    { "id": "studio", "label": "Studio", "credits": 2, "description": "Premium, high-detail generation.",
      "available": false, "reason": "daily_capacity", "availableAt": "2026-08-18T10:00:00.000Z" },
    { "id": "vivid",  "label": "Vivid",  "credits": 2, "description": "Premium photorealistic generation.",
      "available": true,  "reason": null,             "availableAt": null }
  ],
  "aspectRatios": ["16:9","4:3","1:1","3:4","9:16"], "maxCount": 4,
  "sizes": ["original","medium","small","thumb"] }

available tells you whether the model can be used right now; reason is daily_capacity (back at availableAt, an estimate) or disabled.

Check model availability

A tiny, unauthenticated JSON you can poll from any client or dashboard before sending a request. Same available/reason/availableAt semantics as /v1/models.

curl -s https://img-gen-api.sikasio.com/v1/availability
{ "models": {
    "flash":  { "available": true,  "reason": null,             "since": null, "availableAt": null },
    "studio": { "available": false, "reason": "daily_capacity", "since": "2026-08-17T14:02:11.000Z", "availableAt": "2026-08-18T10:00:00.000Z" },
    "vivid":  { "available": true,  "reason": null,             "since": null, "availableAt": null } },
  "updatedAt": "2026-08-17T15:00:00.000Z" }

Recommended client pattern: pick your preferred model; if it is unavailable, either fall back to one that is (flash = 1 credit, studio/vivid = 2) or retry after availableAt. Sending anyway returns 503 model_unavailable with the same information and a Retry-After header — it never creates a job or spends credits.

Check usage

Your key's plan, live credit usage across the three windows, and the queue state.

{ "name": "my-app",
  "plan": "starter",
  "limits": { "rpm": 3, "hourlyCredits": 20, "dailyCredits": 60, "monthlyCredits": 400 },
  "used": { "hour": 1, "day": 4, "month": 7 },
  "remaining": { "hour": 19, "day": 56, "month": 393 },
  "queue": { "total": 4, "queued": 0, "running": 1, "done": 3, "failed": 0, "active": 1 } }

Fetch an image

Serves a generated image (the URLs returned by generation). Public, cacheable, and auto-deleted after a retention window (default 7 days) — download anything you need to keep.

Health check

{ "ok": true, "service": "img-gen-api" }

Parameters at a glance

Errors

All errors are JSON: { "error": { "code": "...", "message": "..." } }

Rate limits

Each plan sets a per-minute request limit plus hourly, daily, and monthly credit allowances (see plans & quotas). Only successful generations consume credits — validation errors and failed jobs don't. When you hit a limit you get 429 (rate_limited or quota_exceeded) with a Retry-After header (seconds). Check remaining credits any time via GET /v1/usage.

FAQ

curl -X POST "https://img-gen-api.sikasio.com/v1/generate?wait=true" \
  -H "X-API-Key: igk_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "a neon coffee cup, studio product shot", "model": "flash"}'

Get an API key

Sign up and get a free key instantly — 25 credits a month, no card required. Upgrade to starter or pro from your account whenever you need more.

Get your free API key →

Questions? support@sikasio.com