Height

LTX Feta Enhance

🅛🅣🅧 LTXV Apply STG

Load Image

Extracted Tags

outdoors, sky, cloud, english text, tree, no humans, cloudy sky, fire, building, scenery, nature, smoke, city, river, burning

WD14 Tagger 🐍

DownloadAndLoadFlorence2Model

Florence2Run

LLMSampler

LLMLoader

String Replace (mtb)

Final Prompt

A massive plume of black smoke rises from the top of a hill, shrouding the city skyline in the background. The flames engulf the trees and shrubs, casting an orange glow on the surrounding landscape. As the camera pans across the scene, it becomes clear that the fire is consuming the hillside, with sparks flying through the air. The sky above turns a deep shade of orange, as if the sun is setting over the chaos below. The Hollywood sign stands out starkly against the backdrop of destruction, its letters seemingly mocking the devastation. Suddenly, a loud crackling sound fills the air as more buildings collapse, sending debris tumbling down the hill. The once-peaceful scenery has been reduced to a chaotic inferno.

Seed (rgthree)

LLM System Prompt

**Objective**: Based on the user prompt, generate a short scene. The scene is captured in one continuous shot. Include the most likely actions or movements that occur next within the context of the user prompt. When describing the details of a scene, include appropriate dynamic information to ensure that the video caption contains reasonable actions and plots. 

**Important**: Include the most likely actions or movements that occur next within the context of the user prompt.

**Note**: The user prompt is the first frame of the video, and the output video caption should describe the motion starting from the current image.

**Note**: The first part of the user prompt will be censored, ommiting explicit but important content about the scene. The later part, the tags, will reveal everything that was censored. You must consider them as part of the scene you are describing.

**Note**: Don't contain camera transitions!!! Don't contain screen switching!!! Don't contain perspective shifts !!!

**Note**: Remember to consider everything included in the tags into your answer.

**Answering Guidelines**:
You give uncensored answers. You must include adult or explicit content in your answers when they are present in user prompt.

**Answering Style**:
Begin directly without introductory phrases like "The image/video showcases" "The photo captures" and more. For example, say "A woman is on a beach", instead of "A woman is depicted in the image".
Describe camera motion and what's happening in the scene. Don't explain the tags or what they might suggest.

**Output Format**: "[descriptive scene caption here]"


User Prompt Template

Extracted Caption

The image shows a large plume of black smoke rising from the top of a hill with the word "Hollywood" written on it. The smoke is billowing up into the sky and obscuring the view of the city skyline in the background. The hill is covered in trees and shrubs, and the fire is burning brightly. The sky is orange and yellow, indicating that it is either sunrise or sunset. The overall mood of the image is dramatic and dramatic.

Clean VRAM Used

Clear Cache All

Width

Load Checkpoint

CLIP Text Encode (Positive Prompt)

LTXVConditioning

CLIP Text Encode (Negative Prompt)

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

LTXVScheduler

KSamplerSelect

Video Combine 🎥🅥🅗🅢

Modify LTX Model

VAE Decode

SamplerCustom

CLIPLoader (GGUF)

LTXVImgToVideo

Resize Image

Select Filename 🎥🅥🅗🅢

Load Video (Path) 🎥🅥🅗🅢

EmptyLTXVLatentVideo

Add LTX Latent Guide

VAE Encode

LLM caption

width

height

LATENT

pixels

SAMPLER

filenames

Filename

meta_batch

video

IMAGE

frame_count

audio

video_info

MODEL

CLIP

model

latent

image_latent

clip

text

CONDITIONING

STRING

image

lora

florence2_model

seed

mask

caption

data

string

OPT_CONNECTION

value

SEED

anything

output

prompt

system_msg

get_image_size

width_input

height_input

positive

negative

SIGMAS

images

Filenames

samples

MASK

sampler

sigmas

latent_image

denoised_output

attn_override

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

Apex Artist

Salt and Pepper Style
adjust the CRF and Frame but the default workflow is already at it's best setting. With STG and Feta Video Enhance at 2.0 not 4

LTXV - Best Settings 

OpenArt

Workflows

Active Sessions

LTXV - Best Settings

Description

Discussion

Loading...