A flat (no subgraphs) LTX-2.5 text-to-video-with-audio graph built entirely from nodes that are either built into ComfyUI core or ship in the official ComfyUI-LTXVideo pack — both are things ComfyUI Manager installs/updates, so nothing outside Manager's reach is required.

Install via Manager

1. Open ComfyUI Manager → Update ComfyUI (LTX-2.5's core nodes are a

   recent addition; an outdated ComfyUI will show them as missing).

2. If any LTXV* node still shows red after updating, use **Manager →

   Install Missing Custom Nodes** (or search "ComfyUI-LTXVideo" in the Custom Nodes Manager) to grab the official Lightricks pack.

No other custom node packs are used in this graph.

Get model access

LTX-2.5's weights are hosted in a gated HuggingFace repo. Visit https://huggingface.co/Lightricks/LTX-2.5, accept the license, and wait for access approval — downloads fail silently without it.

Generative AI Resources, Workflows + Freebies

KoFI: https://www.kofi.com/tensorvizion

Patreon: https://www.patreon.com/tensorvizion

Deploy LTX 2.5 with this LoRA on a large GPU:

Runpod: https://runpod.io?ref=1jcw91