Happy Horse 1.1

Happy Horse 1.1 (alibaba/happyhorse-1.1) is an Alibaba video model: text- and frame-to-video with reference images. Great for dynamic scenes. Generation is asynchronous: POST /v1/videos returns a job, then you poll the status and download the mp4.

Parameters

Field	Description
model	alibaba/happyhorse-1.1 (the short happyhorse-1.1 or the alias happyhorse)
prompt	Scene description, required
seconds	Any integer from 3 to 15; defaults to 3
size	A WxH size from the table below; defaults to 1280x720
input_reference	Image-to-video: one image as the starting frame, {"image_url": "data:image/...;base64,..."}
input_references	Reference-to-video: 2 to 9 images as an array, also data: URLs only
seed	Reproducibility seed: an integer in 0…2147483647. The same prompt with the same seed yields the same clip; without it the upstream picks a random one

Fourteen allowed sizes — seven aspect ratios at two resolutions:

Aspect	720p	1080p
1:1	720x720	1080x1080
16:9	1280x720	1920x1080
9:16	720x1280	1080x1920
4:3	960x720	1440x1080
3:4	720x960	1080x1440
21:9	1680x720	2520x1080
9:21	720x1680	1080x2520

A value outside the list returns 400 with the allowed options listed — check the sizes field in the catalog when in doubt.

Examples

Text-to-video

Python

import requests

job = requests.post(
    "https://api.mixen.ai/v1/videos",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={
        "model": "alibaba/happyhorse-1.1",
        "prompt": "A boxer works a punch bag with a flurry of strikes, sweat spraying, rapid cuts between angles",
        "seconds": "6",
        "size": "1280x720",
    },
).json()
print(job["id"], job["status"])

cURL

curl https://api.mixen.ai/v1/videos \
  -H "Authorization: Bearer $MIXEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "alibaba/happyhorse-1.1",
    "prompt": "A boxer works a punch bag with a flurry of strikes, sweat spraying, rapid cuts between angles",
    "seconds": "6",
    "size": "1280x720"
  }'

Image-to-video from a starting frame

The image goes in as a single data: URL — external links are not accepted:

import base64
import requests

with open("hero.jpg", "rb") as f:
    img = "data:image/jpeg;base64," + base64.b64encode(f.read()).decode()

job = requests.post(
    "https://api.mixen.ai/v1/videos",
    headers={"Authorization": f"Bearer {API_KEY}"},
    json={
        "model": "alibaba/happyhorse-1.1",
        "prompt": "The hero bolts and vaults over an obstacle, sharp dynamic motion",
        "seconds": "8",
        "size": "1280x720",
        "input_reference": {"image_url": img},
    },
).json()
print(job["id"], job["status"])

In image modes Happy Horse inherits the frame format of the input image — see the limits below.

Reference-to-video from several images

Two to nine references:

{
  "model": "alibaba/happyhorse-1.1",
  "prompt": "The character from the first frame sprints through the location from the second, camera in pursuit",
  "seconds": "6",
  "size": "1280x720",
  "input_references": [
    "data:image/png;base64,iVBORw0...",
    "data:image/png;base64,iVBORw0..."
  ]
}

From there it is the usual loop: GET /v1/videos/{id} every 5–10 seconds until completed, then GET /v1/videos/{id}/content for the mp4. Full cycle — video guide.

Price

Billing is per second and depends on the resolution: 720p — 13.2089 ₽ ($0.1482) per second, 1080p — 17.086 ₽ ($0.1917). You are charged actual duration × the size’s rate. Current prices — catalog.

Strengths and limits

• Dynamic scenes are the model’s home turf: motion, action, cuts between angles.
• Durations from 3 to 15 seconds — a continuous range, not fixed buttons.
• Seven frame formats, including the cinematic 21:9 and 9:21.
• Image-to-video from a starting frame and reference-to-video from 2–9 images.
• In image modes (starting frame or references) the model inherits the aspect ratio of the input image and ignores the format from size — that is why the web form hides the format picker for Happy Horse in those modes. The selected resolution sets the rate, not the output geometry; the model honors an exact size only in text-to-video.
• The first-to-last frame transition is not supported — only the starting frame can be passed.

All models — in the catalog. The general video workflow — video guide.