Create Generation

Request a new asynchronous AI generation job for an image or video. The API immediately returns a job ID with PENDING status — poll the GET endpoint to retrieve results.

POST/api/generations

Parameters

modelIdstringRequired

The ID of the AI model to use (e.g. sora-2, kling-v2-6, gemini-2.5-flash-image). See the Overview for the full list.

promptstringRequired

The text prompt describing the desired image or video output. Max 2000 characters.

type"image" | "video"Required

Must be exactly 'image' or 'video'. Must match the capabilities of the chosen model.

aspectRatio"16:9" | "9:16" | "1:1"Optional

Desired output aspect ratio. Defaults to 16:9 if omitted.

negativePromptstringOptional

A description of elements to avoid in the generation. Max 1000 characters.

Request Body

Content-Type: application/json

{
  "modelId": "veo-3.1-generate-preview",
  "prompt": "A futuristic city with flying cars.",
  "type": "video",
  "aspectRatio": "16:9",
  "negativePrompt": "dark, blurry, text"
}

Responses

202Accepted — job successfully enqueued.

{
  "jobId": "cmr1aot6r000004i87w38n1f6",
  "status": "PENDING"
}

400Bad Request — validation failed.

{
  "error": "Validation failed",
  "details": [
    {
      "code": "invalid_type",
      "expected": "string",
      "received": "undefined",
      "path": [
        "prompt"
      ],
      "message": "Required"
    }
  ]
}

401Unauthorized — missing or invalid API key.

{
  "error": "Invalid or missing API key"
}

402Payment Required — insufficient credit balance.

{
  "error": "Insufficient credits to perform this generation"
}

curl -X POST https://editlyplatform-ed06row0u-editly.vercel.app/api/generations \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "modelId": "sora-2",
    "prompt": "A highly detailed majestic lion roaring in the savanna, cinematic lighting.",
    "type": "video",
    "aspectRatio": "16:9"
  }'