Sora 2

Documentation Index

Fetch the complete documentation index at: https://apixo.ai/docs/llms.txt

Use this file to discover all available pages before exploring further.

​

What this model is for

​

Model ID, endpoint, and auth

• Model ID: sora-2
• Base URL: https://api.apixo.ai/api/v1

​

Authentication

Authorization: Bearer YOUR_API_KEY

​

Request Body

{
  "request_type": "async",
  "callback_url": "https://...",
  "provider": "official",
  "input": {
    "mode": "text-to-video",
    "prompt": "...",
    "duration": 12,
    "aspect_ratio": "landscape",
    "image_urls": ["..."],
    "remove_watermark": true
  }
}

​

Parameters

• text-to-video — Generate video from text
• image-to-video — Generate video from image

​

Copy-paste quickstart

curl -X POST "https://api.apixo.ai/api/v1/generateTask/sora-2" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "async",
    "provider": "official",
    "input": {
      "mode": "text-to-video",
      "prompt": "a cinematic flyover of a futuristic city at sunrise, golden hour lighting, slow camera movement revealing towering skyscrapers with holographic advertisements",
      "duration": 12,
      "aspect_ratio": "landscape",
      "remove_watermark": true
    }
  }'

curl -X POST "https://api.apixo.ai/api/v1/generateTask/sora-2" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "request_type": "callback",
    "callback_url": "https://your-server.com/callback",
    "provider": "official",
    "input": {
      "mode": "image-to-video",
      "prompt": "extend this scene into a smooth camera pan with dramatic lighting, the character slowly turns to face the camera with emotional expression",
      "duration": 8,
      "aspect_ratio": "portrait",
      "image_urls": ["https://example.com/ref.jpg"],
      "remove_watermark": true
    }
  }'

​

Response and result format

​

POST /api/v1/generateTask/sora-2

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678"
  }
}

{
  "code": 400,
  "message": "Insufficient credits",
  "data": null
}

​

GET /api/v1/statusTask/sora-2

curl -X GET "https://api.apixo.ai/api/v1/statusTask/sora-2?taskId=task_12345678" \
  -H "Authorization: Bearer YOUR_API_KEY"

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678",
    "state": "success",
    "resultJson": "{\"resultUrls\":[\"https://r2.apixo.ai/video.mp4\"]}",
    "createTime": 1767965610929,
    "completeTime": 1767965652317,
    "costTime": 41388
  }
}

{
  "code": 200,
  "message": "success",
  "data": {
    "taskId": "task_12345678",
    "state": "failed",
    "failCode": "CONTENT_VIOLATION",
    "failMsg": "Content does not meet safety guidelines"
  }
}

​

Polling result fields

​

Polling and webhook result retrieval

​

Common errors

​

Rate Limits

​

Tips

• Generation time:
  • 4-second video: ~2-4 minutes
  • 8-second video: ~3-5 minutes
  • 12-second video: ~4-6 minutes
  • Sora 2 generation takes longer, always use callback mode
  • Submit task, wait 180 seconds, then poll every 10 seconds
• Cinematic quality: Sora 2 is known for ultra-high quality and cinematic narrative capabilities, one of the most advanced video generation models.
• Callback mode: Due to long generation times (several minutes), strongly recommend using callback mode.
• Video expiration: Result URLs are valid for 15 days. Download promptly.
• Content moderation: Prompts must comply with content safety guidelines.
• Aspect ratio selection:
  • landscape: Landscape (default), for traditional video platforms, films, desktop viewing
  • portrait: Portrait, for short video platforms, mobile viewing, social media
• Watermark handling:
  • remove_watermark: true: Auto-remove watermark (default, highly recommended)
  • remove_watermark: false: Keep original watermark
• Long prompts: Sora 2 supports up to 10000-character prompts with exceptional understanding for complex narratives and detailed descriptions.
• Prompt best practices:
  • Camera description: Use film terminology (e.g., “slow dolly in”, “orbiting shot”, “aerial view”, “handheld”)
  • Lighting effects: Describe lighting and atmosphere (e.g., “golden hour”, “dramatic lighting”, “soft diffused light”, “backlit silhouette”)
  • Emotional tone: Convey mood (e.g., “warm”, “tense”, “epic”, “dreamlike”)
  • Action details: Describe character/object movements (e.g., “slowly turns”, “flowing hair”, “swaying leaves”)
  • Scene transitions: Describe how shots unfold (e.g., “zoom from wide to close”, “descend through clouds”)
• Image-to-video:
  • Extends a single image into a complete video
  • Reference image becomes start frame or key frame
  • Describe desired animation effects, camera movement, and narrative development in prompt
  • Ideal for character animation, scene expansion, bringing still images to life
• Duration selection:
  • 4 seconds: Fast preview and quick iteration
  • 8 seconds: Balanced cost and narrative length
  • 12 seconds: Richer scene progression and storytelling
• Image formats: image_urls supports JPG, PNG, WebP, max 10MB per image.
• Professional use:
  • Ideal for movie trailers, commercials, music videos, proof of concept
  • Supports complex narratives and character interactions
  • Can generate coherent multi-shot scenes
• Performance tips:
  • Avoid submitting many tasks during peak hours
  • Clear, structured prompts improve success rates
  • Test with 4-second first, then scale to 8-second or 12-second once satisfied

​

Related links

• Pricing
• Error Codes