Video Generation

Generate short video clips from text prompts, on the generative backend.

Active public models (text-to-video)

Model	Endpoint	Source repo	Notes
Wan2.2 (TI2V-5B)	wan22-video	Wan-AI/Wan2.2-TI2V-5B-Diffusers	Text-to-video
Text-to-Video MS 1.7B	hf-ali-vilab-text-to-video-ms-1.7b	ali-vilab/text-to-video-ms-1.7b	ready

All active endpoints are currently free. Verify in the live catalog:

curl "https://api.colabhive.com/api/builder/v1/endpoints?visibility=public&task_type=text-to-video"

Image-to-video is not active

The Stable Video Diffusion endpoint (hf-stabilityai-stable-video-diffusion-img2vid-xt) is deprecated / inactive — do not rely on it. For image-to-video today, import a model from HuggingFace.

Quick start

from colabhive import ColabHive

client = ColabHive(api_key="hive_...", account_id="YOUR_ACCOUNT_ID")

result = client.endpoints.infer(
    endpoint_id="wan22-video",   # SDK resolves the name to a UUID
    input_data={
        "prompt": "A timelapse of clouds moving over a mountain",
        "num_frames": 25,
        "fps": 7,
    },
)
for art in result.get("result", {}).get("output_artifacts", []):
    client.endpoints.download_artifact(art["url"], f"./{art['filename']}")

Video generation is compute-heavy and may cold-start; poll GET /tasks/{task_id} if the call returns {"status": "queued"}.

Input parameters (text-to-video)

Parameter	Type	Default	Description
prompt	string	required	Description of the video
num_frames	integer	25	Number of frames
fps	integer	7	Frames per second
width	integer	512	Video width
height	integer	512	Video height

Output format

{
  "output_artifacts": [
    {
      "filename": "video.mp4",
      "url": "https://storage.colabhive.com/...",
      "content_type": "video/mp4",
      "size_bytes": 10485760,
      "modality": "video"
    }
  ],
  "metadata": { "fps": 7, "num_frames": 25 }
}

Related

• Generative Models · Image · Audio
• Inference API

---

Authors: José Luis Minich, Maximiliano Lucius.