Video and post-production teams wanting an open 4K model with sound get free weights below $10 million revenue, and a pricier API.

Kenning's take

What it does

LTX-2.5 is an open-weight video model from Lightricks that makes video with sound from text, images or audio, up to 4K in vertical or horizontal formats. It can generate several connected shots in one go while keeping characters, lighting and voice consistent, can pick the clip length itself, and supports HDR output for professional finishing. It can be downloaded from Hugging Face and run locally or in ComfyUI, or used through the LTX API in Fast and Pro versions.

• Native multi-shot generation
• Automatic clip duration
• Up to 4K, vertical and horizontal
• Native HDR and EXR workflows
• Gemma 4 12B text encoder with prompt enhancer
• Pre-trained foundation for fine-tuning and LoRAs

Who it suits. Video creators, VFX and post-production teams, and developers building on an open video model.

Key facts

Category

AI models

Part of

LTX Studio

Made by

Lightricks

Released

11 August 2026

How to get it

open weights on Hugging Face and in ComfyUI; LTX API as ltx-2-5-fast and ltx-2-5-pro; free for organisations under $10M annual recurring revenue

Works on

API, self-hosted, ComfyUI

In the UK

Yes

Free plan

open weights free under $10M annual recurring revenue

API price

ltx-2-5-fast $0.09/s at 720p, $0.13/s 1080p, $0.30/s 4K; ltx-2-5-pro $0.12/s 720p, $0.17/s 1080p, $0.39/s 4K

FAQ

Facts from the maker's own pages and the sources listed. Checked 2026-10-11.