awesome-genmedia Ovi | Text to Video
Generate videos with audio from text descriptions using the Ovi model. Supports multiple resolutions, configurable inference steps, and separate negative prompts for video and audio quality control.
Quick Start
Requires an each::labs API key. Get one at eachlabs.ai.
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "ovi-text-to-video",
"version": "0.0.1",
"input": {
"prompt": "A jazz musician playing saxophone in a dimly lit club, smooth camera movement",
"resolution": "992x512",
"num_inference_steps": 30
}
}'
Parameters
| Parameter | Type | Default | Description |
|---|---|---|---|
| audio_negative_prompt | string | robotic, muffled, echo, distorted | Negative prompt for audio generation |
| negative_prompt | string | jitter, bad hands, blur, distortion | Negative prompt for video generation |
| num_inference_steps | integer | 30 | The number of inference steps |
| prompt | string | The text prompt to guide video generation | |
| resolution | string | 992x512 | Resolution in WxH format. enum: 512x992, 992x512, 960x512, 512x960, 720x720, 448x1120, 1120x448 |
| seed | string | Random seed for reproducibility |
Examples
Widescreen cinematic scene:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "ovi-text-to-video",
"version": "0.0.1",
"input": {
"prompt": "An epic battle scene in a fantasy kingdom with dragons flying overhead and armies clashing",
"resolution": "1120x448",
"num_inference_steps": 40,
"negative_prompt": "jitter, bad hands, blur, distortion, low resolution"
}
}'
Vertical video with seed:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "ovi-text-to-video",
"version": "0.0.1",
"input": {
"prompt": "A dancer performing contemporary dance in a studio with dramatic side lighting",
"resolution": "512x992",
"num_inference_steps": 30,
"seed": "12345"
}
}'
Related Models
- ovi-image-to-video - Image to video with Ovi