Video 1.0 reference_audio_urls: why audio alone is rejected
In Video 1.0, reference_audio_urls (1 to 3 URLs) requires at least one reference image or video. Field limits for images, videos and audio references.

On POST /v1/video-1.0/generate, reference_audio_urls needs at least one reference image or reference video in the same request, so an audio-only reference set is not a valid request. The field takes 1 to 3 audio URLs.
What are the reference limits?
The Video 1.0 request table gives a count for each reference field. All URLs must be public HTTPS.
| Field | Count | Rule |
|---|---|---|
| reference_image_urls | 1 to 9 | Image URLs |
| reference_video_urls | 1 to 3 | Video URLs |
| reference_audio_urls | 1 to 3 | Requires at least one image or video reference |
How do I send audio with a reference image?
Add the audio URL next to an image reference. The example uses placeholder URLs.
curl -X POST https://api.sume.com/v1/video-1.0/generate \
-H "Authorization: Bearer $SUME_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: v1-audio-001" \
-d '{"prompt":"The singer performs to camera","reference_image_urls":["https://example.com/singer.png"],"reference_audio_urls":["https://example.com/song.mp3"]}'Is this the same on /v1/videos?
There, supported_input_references per model says which types (image_url, video_url, audio_url) a model accepts. Audio and video references are honored by the Seedance 2.x models, Wan 3.0, MiniMax H3 and MiniMax H3 Max; Gemini Omni Flash 1.1 accepts video references but not audio.
Where can I read the full field list?
The Video 1.0 docs list every field, and the video generation docs cover the newer route.
What does the reference video field allow?
reference_video_urls takes 1 to 3 video URLs and reference_image_urls takes 1 to 9 images, so a request can combine a few clips with images and then add audio.
On the Video Router side, gemini-omni-flash-1.1 takes up to 10 reference images and up to 3 reference videos of at most 3 seconds each, and has no reference_audio_urls. Read the capabilities for the model you pin.
Sources
Related posts
More in Developers
- routing_preset kling or grok on Video 1.0: accepted, then ignored
Video 1.0 still accepts the retired routing_preset values cost, speed, quality, grok and kling, but all of them run through Auto. How to pick a family instead.
- Video analyses API returns 410 video_analysis_retired: fix
POST /v1/video-analyses answers 410 video_analysis_retired in Sume's dest environment. Stored rows stay readable; new clip inspection uses video inspect.
- Video analysis limits: max_scenes 2-40 and the 90 s and 300 s caps
Sume's video analysis takes max_scenes from 2 to 40 (default 24), warns past about 90 seconds and rejects past 300. Cost, runtime and polling advice.
- What a Sume video analysis returns for each scene
Each scene in a stored video analysis has start and end seconds, summary, visual, shot type, on-screen text, confidence and keyframes. Field-by-field list.
Written by Sume