LTX-Video
Generate videos from text prompts or input images with configurable aspect ratio, resolution, and inference steps. LTX-Video works best with long, descriptive prompts.
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": "ltx-video",
"version": "0.0.1",
"input": {
"prompt": "best quality, 4k, HDR, a tracking shot of a beautiful coastal scene with waves rolling onto sandy beach at golden hour",
"aspect_ratio": "16:9",
"length": 97,
"steps": 30,
"cfg": 3
}
}'
Parameters
| Parameter |
Type |
Default |
Description |
| aspect_ratio |
string |
3:2 |
Output aspect ratio. enum: 1:1, 1:2, 2:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 9:21, 21:9 |
| cfg |
number |
3 |
How strongly the video follows the prompt |
| image |
string |
- |
Optional input image to use as the starting frame |
| length |
integer |
97 |
Number of frames |
| negative_prompt |
string |
low quality, worst quality, deformed, distorted |
Things you do not want to see in your video |
| prompt |
string |
best quality, 4k, HDR... |
Text prompt for the video. Works best with long descriptive prompts |
| seed |
integer |
- |
Set a seed for reproducibility. Random by default |
| steps |
integer |
30 |
Number of steps |
| target_size |
integer |
640 |
Target resolution size |
Examples
Landscape video with image input:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "ltx-video",
"version": "0.0.1",
"input": {
"prompt": "best quality, 4k, HDR, a slow panning shot across a misty mountain valley with sunlight breaking through clouds, cinematic lighting",
"image": "https://example.com/mountain-scene.jpg",
"aspect_ratio": "21:9",
"steps": 40,
"cfg": 3.5
}
}'
Portrait-style video:
curl -X POST https://api.eachlabs.ai/v1/prediction \
-H "Content-Type: application/json" \
-H "X-API-Key: $EACHLABS_API_KEY" \
-d '{
"model": "ltx-video",
"version": "0.0.1",
"input": {
"prompt": "best quality, 4k, a woman walking through a neon-lit Tokyo street at night, rain reflections on the ground, cyberpunk atmosphere",
"aspect_ratio": "9:16",
"length": 97,
"steps": 30,
"negative_prompt": "low quality, blurry, distorted faces, ugly",
"seed": 12345
}
}'
Related Models
- mochi-1 - High-quality video generation with frame control
- flux-dev - Text-to-image generation with Flux
- toolkit - Video format conversion and processing
Documentation
1---2name: ltx-video3description: LTX-Video | AI Video Generation. Generate videos from text prompts with aspect ratio and resolution control. Triggers: ltx, ltx-video, text to video, video generation, ai video4---56# LTX-Video78Generate videos from text prompts or input images with configurable aspect ratio, resolution, and inference steps. LTX-Video works best with long, descriptive prompts.910## Quick Start1112> Requires an each::labs API key. Get one at [eachlabs.ai](https://eachlabs.ai).1314```bash15curl -X POST https://api.eachlabs.ai/v1/prediction \16 -H "Content-Type: application/json" \17 -H "X-API-Key: $EACHLABS_API_KEY" \18 -d '{19 "model": "ltx-video",20 "version": "0.0.1",21 "input": {22 "prompt": "best quality, 4k, HDR, a tracking shot of a beautiful coastal scene with waves rolling onto sandy beach at golden hour",23 "aspect_ratio": "16:9",24 "length": 97,25 "steps": 30,26 "cfg": 327 }28 }'29```3031## Parameters3233| Parameter | Type | Default | Description |34|-----------|------|---------|-------------|35| aspect_ratio | string | 3:2 | Output aspect ratio. enum: 1:1, 1:2, 2:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 9:21, 21:9 |36| cfg | number | 3 | How strongly the video follows the prompt |37| image | string | - | Optional input image to use as the starting frame |38| length | integer | 97 | Number of frames |39| negative_prompt | string | low quality, worst quality, deformed, distorted | Things you do not want to see in your video |40| prompt | string | best quality, 4k, HDR... | Text prompt for the video. Works best with long descriptive prompts |41| seed | integer | - | Set a seed for reproducibility. Random by default |42| steps | integer | 30 | Number of steps |43| target_size | integer | 640 | Target resolution size |4445## Examples4647**Landscape video with image input:**48```bash49curl -X POST https://api.eachlabs.ai/v1/prediction \50 -H "Content-Type: application/json" \51 -H "X-API-Key: $EACHLABS_API_KEY" \52 -d '{53 "model": "ltx-video",54 "version": "0.0.1",55 "input": {56 "prompt": "best quality, 4k, HDR, a slow panning shot across a misty mountain valley with sunlight breaking through clouds, cinematic lighting",57 "image": "https://example.com/mountain-scene.jpg",58 "aspect_ratio": "21:9",59 "steps": 40,60 "cfg": 3.561 }62 }'63```6465**Portrait-style video:**66```bash67curl -X POST https://api.eachlabs.ai/v1/prediction \68 -H "Content-Type: application/json" \69 -H "X-API-Key: $EACHLABS_API_KEY" \70 -d '{71 "model": "ltx-video",72 "version": "0.0.1",73 "input": {74 "prompt": "best quality, 4k, a woman walking through a neon-lit Tokyo street at night, rain reflections on the ground, cyberpunk atmosphere",75 "aspect_ratio": "9:16",76 "length": 97,77 "steps": 30,78 "negative_prompt": "low quality, blurry, distorted faces, ugly",79 "seed": 1234580 }81 }'82```8384## Related Models8586- [mochi-1](../mochi-1/) - High-quality video generation with frame control87- [flux-dev](../flux-dev/) - Text-to-image generation with Flux88- [toolkit](../toolkit/) - Video format conversion and processing8990## Documentation9192- [each::labs Docs](https://docs.eachlabs.ai)93- [API Reference](https://docs.eachlabs.ai/api/overview)
awesome-genmedia