Skip to content
WorkflowListedExternal
Download

LTX 2.3 - Unlimited Video Length - Sing or Talk (Alternating Shots) — v7.0

Made changes to workflow... add convrot instead of gguf for better reliability

Sourcecivitai.com/models/2422765
VersionRelease 3368860 — LTX 2.5 Long Speaker and Singer.jsonEstimated VRAMNot yet characterizedUnratedgrade

Runtime profile

Access
External workflow
Status
Listed
Version
LTX 2.5 Long Speaker and Singer.json
Source
Civitai
Updated
Nov 3, 2026

What you need

Model files
3.1 GiB known subtotal

Remaining hardware or download-size estimates are not available yet.

Requirements

Models

7

Node packs

15

  • basic_data_handling

    Not resolved

    Node pack · Registry

  • comfy-mtb

    Found

    Node pack · Registry

  • comfyui-easy-use

    Found

    Node pack · Registry

  • ComfyUI-GGUF

    Possible

    Node pack · Registry

  • comfyui-gguf-reboot

    Not resolved

    Node pack · Registry

  • comfyui-handynodes-kt

    Not resolved

    Node pack · Registry

  • comfyui-kjnodes

    Found

    Node pack · Registry

  • ComfyUI-LTXVideo

    Found

    Node pack · Registry

  • ComfyUI-MelBandRoFormer

    Found

    Node pack · Registry

  • comfyui-videohelpersuite

    Found

    Node pack · Registry

  • GetNode

    Not resolved

    Node pack · Unknown

  • kijai/ComfyUI-KJNodes

    Found

    Node pack · GitHub · kijai/ComfyUI-KJNodes

  • Label (rgthree)

    Not resolved

    Node pack · Unknown

  • rgthree-comfy

    Found

    Node pack · Registry

  • SetNode

    Not resolved

    Node pack · Unknown

AI-generated commentary

AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.

Purpose

  • This ComfyUI workflow is for making long videos of someone talking or singing, with alternating shots; its listed nodes include prompt, image, audio, and video inputs plus video assembly.

Flow

  • Its node list includes loop controls, prompt looping, audio chunking, trimming, and video assembly, suggesting a multi-step approach to longer videos.

    Suggestion · not verified

Inputs and outputs

  • Named inputs include text prompts, multiple images, uploaded audio, or a video path; a VHS_VideoCombine node is listed for video assembly.

Setup

  • The creator says the workflow requires their custom nodes, called TK Handy Nodes.

Dependencies

  • Listed node packs: basic_data_handling, comfy-mtb, comfyui-easy-use, ComfyUI-GGUF, comfyui-gguf-reboot, comfyui-handynodes-kt, comfyui-kjnodes, ComfyUI-LTXVideo, ComfyUI-MelBandRoFormer, comfyui-videohelpersuite, and rgthree-comfy.

  • Listed model files: gemma4-12b-with-proj-ltx-2.5-comfy-int8-convrot.safetensors, gemma4-12b-with-proj-ltx-2.5-Q5_K_M.gguf, ltx-2-spatial-upscaler-x2-1.0.safetensors, ltx-2.5-audio-vae-bf16.safetensors, LTX-2.5-Distilled-Q4_K_M.gguf, ltx-2.5-video-vae-bf16.safetensors, and MelBand

Configuration

  • If you use lip sync often, consider the creator’s suggestion to use LTX 2.5; they say this update uses a Convrot Gemma text encoder.

    Suggestion · not verified

Cautions

  • Check which LTX version you need before downloading: the title says LTX 2.3, while the description says it was updated for 2.5.

    Suggestion · not verified

  • No usage license is listed here; check the source page before reusing or sharing the workflow.

    Suggestion · not verified

Does this need to be edited?

Sign in to send an edit request.

Sources

1 source
CivitaiPrimaryVersion 3368860
civitai.com/models/2422765

Verified Nov 3

Workflow instructions

Author notes from LTX 2.5 Long Speaker and Singer.json · selected release

___________ .__ ____ __. .__ .__ __ \__ ___/___________ _____| |__ | |/ _|____ | | | | ____ _____/ |_ ___________ | | \_ __ \__ \ / ___/ | \| < / _ \| | | | _/ __ \_/ ___\ __\/ _ \_ __ \ | | | | \// __ \_\___ \| Y \ | ( <_> ) |_| |_\ ___/\ \___| | ( <_> ) | \/ |____| |__| (____ /____ >___| /____|__ \____/|____/____/\___ >\___ >__| \____/|__| \/ \/ \/ \/ \/ \/ **************LTX 2.5 VERSION************* VAE https://huggingface.co/Lightricks/LTX-2.5/tree/main/vae TEXT ENCODER (convrot) https://huggingface.co/Lightricks/LTX-2.5/tree/main/text_encoders UPSCALER https://huggingface.co/Lightricks/LTX-2.5/tree/main/latent_upscale_models GGUF https://huggingface.co/Abiray/LTX-2.5-Distilled-GGUF/tree/main GEMMA https://huggingface.co/Lightricks/LTX-2.5/tree/main/text_encoders

Download files

Model artifacts

2 artifacts

ltx25UnlimitedVideoLength_v70.zip

zip · 23.9 KB · SHA-256 1c06d38b8bc5…801a · Civitai

Download file

LTX 2.5 Long Speaker and Singer.json

json · 131 KB · SHA-256 d3bf754093b4…f968

Download file

Source excerpts

2 excerpts
NoteNov 3civitai.com/models/2422765

Source context: 1936 downloads · Type Workflows · Base model LTXV 2.5

NoteOct 2civitai.com/models/2422765

Source context: 1602 downloads · Type Workflows · Base model LTXV 2.5