Abra Video Generation

Abra is another video generation line, offering both text-to-video (t2v) and reference-image-to-video (r2v) modes, with 8s / 10s durations and optional landscape or portrait aspect ratios. Requests use the unified asynchronous protocol; see Video Generation (Asynchronous Protocol).

Models on Sale

Mode	Model
Text-to-video	abra_t2v_8s_landscape, abra_t2v_8s_portrait, abra_t2v_10s_landscape, abra_t2v_10s_portrait
Reference-image-to-video	abra_r2v_8s_landscape, abra_r2v_8s_portrait, abra_r2v_10s_landscape, abra_r2v_10s_portrait

Submission Example

curl https://api.4allapi.com/v1/video/generations \
  -H "Authorization: Bearer $AIPROXY_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "abra_r2v_10s_portrait",
    "prompt": "The character in the reference image turns around and smiles under the cherry blossom tree, with the camera slowly pushing in",
    "image": "https://your-cdn.example.com/character.png"
  }'

Usage Notes

• For r2v, the reference image determines the consistency of the character/subject; a clear front-facing image works best;
• 8s and 10s have different unit prices; before batch usage, check the pricing in the Model Marketplace;
• For comparisons with other series, see Image and Video Generation Models.