Steps

Florence Text (Auto generated)

The video shows a large house on fire. The house is a two-story building with a sloping roof and a chimney on the right side. The front of the house is engulfed in flames, with thick black smoke billowing up into the sky. The flames are bright orange and yellow, and the house appears to be completely destroyed. The sky is dark and cloudy, and there are trees in the background, suggesting that the video was taken at night. The ground is covered in fallen leaves and there is a path leading up to the house. The overall mood of the video is eerie and foreboding.

Pre Text

After text

Find and Replace

Model

Florence2Run

Text Concatenate

replace "photo/image" caption with

VAE Encode

LTX Attention Override

Modify LTX Model

LTX Apply Perturbed Attention

Add LTX Latent Guide

Resize Image

EmptyLTXVLatentVideo

Final Frame Selector

Preview Image / Last Frame

Extend Video = True

SALT  -   (CRF)

Your own Prompt

Switch to own prompt = TRUE

PEPPER  -  (Frame Rate)

Seed Generator

Dimension (default 768)

Unet Loader (GGUF)

Load VAE

CLIPLoader (GGUF)

CLIP Text Encode (Positive Prompt)

LTXVConditioning

CLIP Text Encode (Negative Prompt)

low quality, worst quality, deformed, distorted, disfigured, motion smear, motion artifacts, fused fingers, bad anatomy, weird hand, ugly

LTXVScheduler

SamplerCustom

KSamplerSelect

Load Image Input

Video Combine 🎥🅥🅗🅢

VAE Decode

Select Filename 🎥🅥🅗🅢

Load Video (Path) 🎥🅥🅗🅢

Video Length

LTX/STG

Loader/sampler/misc

Extend: Last Frame current clip

GUI Image to Video with STG and Florence Autocaption  GGUF V6

Make it move       (see note on howto)

Prepare Input Image

Florence Caption

text

replace

result_text

replacement_count_number

replacement_count_float

replacement_count_int

lora

florence2_model

filenames

Filename

meta_batch

video

IMAGE

frame_count

audio

video_info

image

get_image_size

width_input

height_input

width

height

seed

mask

caption

data

model

attn_override

MODEL

FLOAT

STRING

images

Filenames

on_true

on_false

ATTN_OVERRIDE

length

LATENT

latent

image_latent

pixels

steps

SIGMAS

positive

negative

sampler

sigmas

latent_image

noise_seed

output

denoised_output

SAMPLER

samples

text_a

text_b

text_c

text_d

clip

CONDITIONING

MASK

CLIP

frame_rate

Load Image Here

git+https://github.com/WASasquatch/img2texture.git@d6159abea44a0b2cf77454d3d46962c8b21eb9d3

Apex Artist

<p>LTXV Salt n Pepper HD</p><p>Tutorial video how to use it : &nbsp;<a href="https://youtu.be/ilQOZkWAtEU" target="_blank">https://youtu.be/ilQOZkWAtEU</a> </p>

LTXV Salt n Pepper HD

OpenArt

Workflows

Active Sessions

LTXV Salt n Pepper HD

Description

Discussion

Loading...