# LTX Open-Sources AI Video Generation Foundation Model LTX-2.5

- Category: Flash (快讯)
- Published: 2026-08-13T11:48:00.000Z
- Source date: 2026-08-13
- HTML: https://www.88z.ai/en/news/flash-20260813-ff254974273f0f17
- Markdown: https://www.88z.ai/en/news/flash-20260813-ff254974273f0f17.md
- Other language: https://www.88z.ai/news/flash-20260813-ff254974273f0f17.md

Open-world model company LTX has released LTX-2.5 with native ComfyUI integration. On dual NVIDIA GB200 systems, it generates a 10-second 720P video in just 6.8 seconds, while the API version outputs 1080P video in 23.7 seconds. The new model offers improved pixel fidelity, multi-shot consistency, and diffusion rendering quality, accessible via Hugging Face, ComfyUI, and API.

## At a glance

- LTX-2.5 natively integrates with ComfyUI, generating 720P video in 6.8 seconds on dual GB200s
- Improved pixel fidelity and multi-shot consistency, available via Hugging Face and API

## The update

Open-world model company LTX has officially launched LTX-2.5, its latest AI video generation foundation model, featuring native integration with ComfyUI. In terms of performance, generating a 10-second 720P video takes only 6.8 seconds on two NVIDIA GB200 units, while outputting a 1080P video via the API requires 23.7 seconds. LTX-2.5 delivers significant visual improvements, including higher pixel fidelity, enhanced consistency across multiple shots, and advanced diffusion fidelity rendering technology. Developers and users can currently access and utilize the model through the Hugging Face platform, ComfyUI workflows, and the official API interface.

## Sources

- [IT之家](https://www.ithome.com/0/988/766.htm) (2026-08-13)

This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.