Skip to content

Single lookup

Generate an H3 video from text or frame images

An AI-generated video file, optionally using first and last frame images

Run it.From your AI, or try it here.

Use Ahel to generate an H3 video from text or frame images. Prompt: <prompt> Model: <model> Duration: <duration> Resolution: <resolution>

Try it

Paste this into ChatGPT or Claude connected to Ahel.

You get: an AI-generated video file, optionally using first and last frame images.

For developers
Action id
video-gen.h3.generate
Inputs
  • prompt: What the video shows, up to 2,000 characters.
  • model: hailuo-3 (H3: 2K, with audio) or hailuo-3-max (H3 Max: 480p or 768p).
  • duration: Length in seconds, 5 to 15. The price is per second.
  • resolution: 2K for hailuo-3; 480p or 768p for hailuo-3-max.
  • aspect_ratio (optional): Frame shape; 16:9 when left out.
  • image_url (optional): Public https URL of an image (JPEG, PNG or WebP) used as the first frame. The provider fetches it.
HTTP call
curl -s https://ahel.ai/api/platform/v2/actions/video-gen.h3.generate/runs \
  -H "Authorization: Bearer $AHEL_TOKEN" -H "Content-Type: application/json" \
  -d '{"inputs":{"prompt":"<prompt>","model":"<model>","duration":"<duration>","resolution":"<resolution>"},"max_cost_usd":"<your cap>"}'
API reference

More on AI Generation.Other single lookups.

Not affiliated with the platforms named. Names and logos belong to their owners.