Unlimited Open Source Models

Get Plan
Skip to main content
Available now on ModelsLab · Video Generation

LTX 2.3 ProAPI for Text and Image to Video with Audio

Part of the LTX 2.3 API family on ModelsLab.

Try LTX 2.3 Pro Text To VideoAPI Documentation

Generate Precise Videos

1080p, 1440p or 4K

Three resolutions on one endpoint

Set resolution to 1920x1080, 2560x1440 or 3840x2160, at 25 or 50 fps. 1440p costs 2x and 4K costs 4x the 1080p rate.

6, 8 or 10 seconds

You pick the clip length

Price follows duration: $0.576 for 6 seconds, $0.768 for 8 and $0.96 for 10 at 1080p.

Synced audio

Sound in the same call

Set generate_audio to true for ambient sound and effects that match the scene, or false for a silent clip you score later.

Examples

See what LTX 2.3 Pro Text To Video can create

Copy any prompt below and try it yourself in the playground.

City Timelapse

“Slow aerial pan over futuristic city skyline at dusk, neon lights flicker on skyscrapers, flying cars weave between towers, cinematic lighting, shallow depth of field, 24fps.”

Ocean Waves

“Gentle waves crash on rocky shore at sunrise, golden light reflects on water, seagulls glide overhead, serene atmosphere, wide angle lens, subtle camera push-in, 4K.”

Forest Path

“Tracking shot along misty forest path, sunlight filters through ancient trees, leaves rustle in breeze, volumetric fog, realistic motion, 16:9 aspect ratio.”

Tech Abstract

“Pulsing holographic data streams in dark void, neon blue circuits form geometric patterns, smooth camera orbit, high contrast, sci-fi style, 50fps smoothness.”

For Developers

A few lines of code.
4K video with audio. One call.

ModelsLab handles the infrastructure: fast inference, auto-scaling, and a developer-friendly API. No GPU management needed.

  • Serverless: scales to zero, scales to millions
  • Pay $0.096 per second at 1080p, on plans from $21/month
  • Python and JavaScript SDKs, plus REST API
API Documentation 
import requests


response = requests.post(
    "https://modelslab.com/api/v7/video-fusion/text-to-video",
    json={
  "fps": "25",
  "key": "YOUR_API_KEY",
  "prompt": "Slow aerial pan over a futuristic city skyline at dusk, neon lights flicker on skyscrapers, cinematic lighting, shallow depth of field",
  "duration": "6",
  "model_id": "ltx-2-3-pro-t2v",
  "resolution": "1920x1080",
  "generate_audio": true
}
)
print(response.json())

FAQ

Common questions about LTX 2.3 Pro

Read the docs

LTX 2.3 Pro costs $0.096 per second of video at 1080p. 1440p is 2x ($0.192 per second) and 4K is 4x ($0.384 per second). A 6-second 1080p clip costs $0.576 and a 10-second 4K clip costs $3.84. Image to video has the same rates.

LTX 2.3 Pro (ltx-2-3-pro-t2v and ltx-2-3-pro-i2v) is the Lightricks-hosted model: 1080p to 4K, 6 to 10 seconds, synced audio, billed per second on every plan. The open LTX 2.3 model (ltx-2.3) runs on ModelsLab GPUs at /api/v6/video/text2video_ultra: 5 to 12 seconds, 1:1, 16:9 or 9:16, from $0.075 per second, and it is included in the $149/month Open Source Unlimited plan.

POST to https://modelslab.com/api/v7/video-fusion/text-to-video with model_id ltx-2-3-pro-t2v, a prompt, resolution, duration and fps. For image to video, POST to https://modelslab.com/api/v7/video-fusion/image-to-video with model_id ltx-2-3-pro-i2v and init_image set to the first frame.

Resolutions 1920x1080, 2560x1440 and 3840x2160. Durations of 6, 8 or 10 seconds. Frame rates of 25 fps (the default) or 50 fps.

Yes. Set generate_audio to true and the MP4 comes back with AI-generated sound that matches the scene. Set it to false for a silent clip.

No. LTX 2.3 Pro runs on the model maker's servers and is billed per second on every plan. The open LTX 2.3 model runs on ModelsLab GPUs and is included in Open Source Unlimited.

Ready to create?

Start generating with LTX 2.3 Pro on ModelsLab.

Try LTX 2.3 Pro Text To VideoAPI Documentation