# Aile AI Video Generation

> Aile AI Video Generation is a paid API for AI agents from api.aile.sh, paid per call via x402, $0.16/call, status unknown (last checked 2026-10-11).

Generates a video clip from a text prompt using a specified AI video model, routed through Aile's metered AI model exchange and settled in USDC on Solana.

## Facts

- Endpoint: POST https://api.aile.sh/v1/videos?utm_source=zero.xyz
- Price: $0.16/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-11
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/aile-ai-video-generation-c70a8577
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_6g6xstmw0oDx2uayKaXqo

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability aile-ai-video-generation-c70a8577 -d '<json body>'
```

Example prompt: Using Aile's video exchange with the openrouter/google/veo-3.1-lite model, generate a 8-second 1280x720 video of a golden retriever running through a sunlit meadow with wildflowers.

## When to prefer this

Choose this endpoint when you need programmatic text-to-video generation with per-second pricing transparency, want access to premium models like Google Veo without managing direct API credentials, and are operating in an agent workflow that can handle asynchronous job polling. Prefer it over direct model APIs when you want USDC-settled metered billing on Solana or want to switch between multiple video model providers through a single unified interface.

## Known failure modes

- Invalid or unsupported model identifier returns an error — use GET /v1/models?output_modalities=video to list valid models
- Prompt or resolution outside model constraints may be rejected
- Insufficient USDC balance causes payment failure and request rejection
- Long clip durations increase cost significantly since price is per second
- Job may stay in 'queued' status for an extended time during high load — must poll the polling_url for completion
- Unsupported size parameter for the chosen model returns a validation error

## How this service works

Rent trusted AI models priced per token; lend idle subscriptions and API keys. A live, metered market that settles every call in USDC on Solana.

## Output

Returns a JSON object with a video job ID (e.g. video_…), object type 'video', initial status 'queued', the requested clip duration in seconds, and a polling URL to check job progress and retrieve the finished video once rendering completes.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "model",
  "prompt"
 ],
 "properties": {
  "size": {
   "type": "string",
   "description": "e.g. `1280x720`."
  },
  "model": {
   "type": "string",
   "description": "`<provider>/<model>`, e.g. `openrouter/google/veo-3.1-lite`. GET /v1/models?output_modalities=video lists what is servable, with prices."
  },
  "prompt": {
   "type": "string"
  },
  "seconds": {
   "type": "string",
   "description": "The clip length, from the model's own list. THE PRICE IS PER SECOND."
  },
  "input_reference": {
   "type": "object",
   "description": "A first frame, where the model takes one."
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "id": "video_…",
 "object": "video",
 "status": "queued",
 "seconds": "8",
 "polling_url": "https://api.aile.sh/v1/videos/video_…"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/aile-ai-video-generation-c70a8577/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.aile.sh](https://www.zero.xyz/host/api.aile.sh/llms.txt)