Video generation models
Openference exposes video models on POST /v1/videos with polling via GET /v1/videos/. Use the model ids below in your request body. Aliases are also accepted (e.g.agnes-video-2.5-flash → Agnes-Video-2.5-Flash).
Agnes Video 2.5 Flash (Agnes-Video-2.5-Flash)
Task: text-to-video, keyframe control, image/audio referenceOutput: 720P MP4, 4–12 seconds
Text-to-video example
video_id from the response, then poll:
Errors
See POST /v1/videos and Common errors for the full list.