LTX 2.3 AI Video

LTX 2.5 AI Video Generator is an independent third-party service built around Lightricks' open-weights LTX 2.5 model. It turns text prompts, uploaded images, or audio into finished video clips with synchronized sound, rendered in the browser on cloud GPUs. Aimed at creators and marketers who need ready-to-post footage without local hardware or editing.

How it works

LTX 2.5 generates video and audio in the same forward pass, so clips come back with ambient sound, music beds, and effects timed to the action. It supports text-to-video, image-to-video, and audio-to-video inputs. The model uses a 22B-parameter Diffusion Transformer with a Gemma 4 text encoder and a diffusion video decoder, enabling detailed 4K output and complex prompt adherence.

Key capabilities

The generator supports 720p, 1080p, 1440p, and 4K (2160p) at 24, 25, 48, or 50 FPS, in landscape (16:9) or native portrait (9:16). Clips can run up to 20 seconds at lower resolutions and frame rates, or up to 10 seconds at higher settings. Auto duration lets the model end the clip where the action ends, and Diffusion Fidelity Rendering allocates extra compute to complex frames.

Who it is for

This service is for creators, marketers, and businesses that need finished video clips without local GPU hardware or editing software. It runs entirely in the browser on cloud GPUs, so users can generate from a laptop, tablet, or phone. The open-weights model is free for individuals and companies under $10M in annual revenue under the LTX-2.x Community License, though generation here runs on the service's infrastructure.

Pricing and limits

The service offers a free tier with 0 credits available, and paid plans are available for additional generation. Pricing details are not fully published on the site, but the free tier allows initial use. The service also includes compliance measures with input/output moderation and blocking controls for restricted model routes.