Live
Wan-2.2 Text-to-Video A14B for AI agents
Open-weights Wan 2.2 (14B): 1 to 10 second clips at 480p to 720p. The cheapest way to try an idea. $0.08 per second, no markup, paid per call from your AMXP balance.
What it costs
| Unit | Price | Applies to |
|---|---|---|
| second | $0.08 per second | Per second of output, counted at 16 frames per second: $0.08 at 720p, $0.06 at 580p, $0.04 at 480p. |
List price, 0% markup. Your agent sees the exact figure from amxp.inspect() before it runs anything; the estimate is held from your balance and only the real cost is settled.
Parameters
What amxp.execute() accepts for wan-2.2/text-to-video. Output: video, typically 40–150 s. $0.08 per second; Per second of output, counted at 16 frames per second: $0.08 at 720p, $0.06 at 580p, $0.04 at 480p.
| Name | Type | Description |
|---|---|---|
prompt required | string | What happens in the video. |
negative_prompt | string | What to avoid. |
num_frames | integer (default 81) | Frames to generate; 16 frames is one second. |
frames_per_second | integer (default 16) | Playback rate of the file. |
resolution | 480p · 580p · 720p (default 720p) | Output resolution. |
aspect_ratio | 16:9 · 9:16 · 1:1 (default 16:9) | Frame shape. |
seed | integer | Fix for repeatable results. |
How to call it
Callable now from the API, the MCP server and the console.
// Wan-2.2 Text-to-Video A14B: check the price, then run it.
const spec = await amxp.inspect("wan-2.2/text-to-video") // parameters and price per second
const result = await amxp.execute("wan-2.2/text-to-video", { prompt: "…" })
// result.output.video is served from amxp.aiAlternatives in Video Generation
Your agent can compare these by fit and price with amxp.discover() and pick per task.