• Picasso AI Logo
    Logo Picasso IA
  • Home
  • AI Image
    Nano Banana 2
  • AI Video
    Veo 3.1 Lite
  • AI Chat
    Gemini 3 Pro
  • Edit Images
  • Upscale Image
  • Remove Background
  • Text to Speech
  • Effects
    NEW
  • Generations
  • Billing
  • Support
  • Account
  1. Collection
  2. Text to Video
  3. Ltx 2.3 Pro

Create 4K AI Videos from Text with LTX 2.3 Pro

LTX 2.3 Pro is a high-fidelity video generation model built to handle the full creative loop, from a written description to a finished clip in minutes. It solves a concrete problem: getting high-quality video output without juggling multiple tools for text input, image animation, audio sync, and post-production edits. Resolutions reach up to 4K at 50 FPS, which means the output is sharp enough for client-facing work, not just quick drafts. The model runs five distinct tasks from a single interface. Text-to-video takes a written prompt and outputs a moving scene, while image-to-video animates a still photo from the first frame forward. Audio-to-video generates visuals driven by a sound file. The retake task lets you replace a specific short segment of an existing video without regenerating the full clip. Extend adds new footage to the start or end of a video, letting you stretch a scene rather than rebuild it. In practice, a content creator might draft a product clip from a prompt, refine the opening shot using retake, then extend the ending to hit a target duration. Camera motion presets, dual frame inputs, and aspect ratio controls give you room to iterate on each variable without starting from scratch. The workflow is built for repetition, which makes it useful for anyone producing a steady stream of short-form video.

Official

Lightricks

1.9k runs

Ltx 2.3 Pro

2026-03-05

Commercial Use

Table of contents

  • Overview
  • How It Works
  • Frequently Asked Questions
  • Credit Cost
  • Features
  • Use Cases
Get Nano Banana Pro

Overview

LTX 2.3 Pro is a multi-mode video generation model available on Picasso IA that produces high-fidelity clips up to 4K at 50 FPS. It accepts text, images, and audio as inputs, so you can start wherever your creative process begins. A marketing team might generate product videos from a written brief, while a filmmaker might animate a storyboard frame and extend it into a full scene. Five distinct generation tasks live inside one tool, which removes the need to move between separate applications for each stage of video production.

How It Works

  • Write a text prompt describing the scene you want, or upload an image to use as the first or last frame of the video.
  • Select the task type: text-to-video, image-to-video, audio-to-video, retake, or extend.
  • Set your resolution (1080p, 2K, or 4K), aspect ratio (16:9 or 9:16), duration (6, 8, or 10 seconds), and frame rate.
  • For retake, enter the start time and segment length of the section you want to replace. For extend, choose whether new footage gets added to the start or end.
  • Generate the video and download the output file directly.

Frequently Asked Questions

Do I need programming skills or technical knowledge to use this? No, just open LTX 2.3 Pro on Picasso IA, adjust the settings you want, and hit generate.

Is it free to try? Yes, you can run LTX 2.3 Pro without a paid subscription to check the results. Credit limits vary by account level, so heavier usage may require a plan upgrade.

How long does it take to get results? Generation time depends on resolution and duration. A 6-second clip at 1080p finishes faster than a 10-second 4K output. Most runs return within a few minutes.

What output formats are supported? LTX 2.3 Pro returns standard video files you can download and bring into any editing software or upload directly to social platforms.

Can I customize the output style or motion? Yes. You control resolution, frame rate, aspect ratio, duration, and camera motion. For image-to-video tasks, setting both a first and last frame defines the motion path between them.

What is the retake feature and when should I use it? Retake isolates a specific segment of a video you already generated and replaces just that portion. It is useful when the overall clip looks right but one short moment needs to be redone.

Where can I use the outputs? Outputs on supported plans come without watermarks, so they work in client presentations, social content, short films, and any other context that calls for original video.

Credit Cost

The credit cost for this model varies based on the settings you choose. Below are the costs per configuration:

ConfigurationCredits
1080p1.6per second
2k3.2per second
4k6.4per second

Features

Everything this model can do for you

Five generation modes

Switch between text-to-video, image-to-video, audio-to-video, retake, and extend from one interface.

4K resolution at 50 FPS

Export high-fidelity video at up to 4K, suitable for professional delivery and broadcast use.

Audio-driven generation

Feed an audio file directly to produce video visuals that follow the tone and rhythm of the track.

Retake editing

Replace a short segment of an existing video without regenerating the entire clip from scratch.

Camera motion presets

Apply dolly, jib, static, or focus-shift motion to any generated scene with a single setting.

Frame interpolation

Set both a first and last frame so the model fills in the motion between them automatically.

Flexible aspect ratios

Output in 16:9 for standard widescreen or 9:16 for vertical social video formats.

Portrait and landscape framing:

Switch between 16:9 and 9:16 aspect ratios to fit YouTube, TikTok, Instagram Reels, or any other format in one click.

Use Cases

Generate a 6-to-10-second cinematic clip from a written scene description, with camera motion like dolly-in or jib-up applied to the output

Animate a still portrait photo into a short video by setting it as the first frame and describing the movement you want

Sync a generated video to an audio track by feeding an audio file directly into the audio-to-video task

Fix a specific 2-second moment in an existing video by isolating that segment with retake instead of regenerating the whole clip

Extend an existing video clip by appending new footage to the start or end while keeping the original content intact

Create a 4K vertical social video at 9:16 aspect ratio from a product description paired with a reference image

Interpolate between a first and last frame image to produce a smooth transitional video without describing every intermediate step

Animate a flat illustration by uploading it as the first frame and describing the motion you want, turning static artwork into a moving asset.

Switch Category

Effects

Text To Image

Text To Image

Text To Video

Large Language Models

Large Language Models

Text To Speech

Text To Speech

Super Resolution

Super Resolution

Lipsync

AI Music Generation

AI Music Generation

Video Editing

Speech To Text

Speech To Text

AI Enhance Videos

Remove Backgrounds

Remove Backgrounds