Load VAE

Nunchaku Text Encoder Loader

Nunchaku FLUX DiT Loader

Nunchaku FLUX.1 LoRA Loader

Preview Image

Empty Latent Image

Negative Cond Placeholder

KSampler (inspire)

CLIP Text Encode (Positive Prompt)

Extreme closeup photograph of a single royal bengal tiger staring directly. The word "Deck Diavalo" is across the lower portion of the image in large white stylized letters, with brush strokes resembling those made with Japanese calligraphy. Natural lighting to capture authentic eye shine and depth.

VAE Decode

MODEL

images

samples

IMAGE

LATENT

CLIP

clip

CONDITIONING

model

positive

negative

latent_image

scheduler_func_opt

git+https://github.com/WASasquatch/img2texture.git@a546609a915caf1dcb84a15bf44b5b639ad924e6

Deck Diavalo

This workflow utilizes the excellent Nunchaku SVDQ Model to with low VRAM settings.The workflow includes multi-LoRA setup if needed. Tested on: RTX 3060 Laptop GPU with 6 GB VRAMTime taken per Iteration: ~1-1.5 s/it8-10 steps are good enough for good output. Checkpoint: <a href="https://huggingface.co/mit-han-lab/nunchaku-flux.1-dev/resolve/main/svdq-int4_r32-flux.1-dev.safetensors" target="_blank">Nunchaku-Flux.1-Dev</a>Check Nunchaku installation instructions at MIT HAN lab github <a href="https://github.com/mit-han-lab/nunchaku" target="_blank">here</a>

Nunchaku Text to Image (Low VRAM)

OpenArt

Workflows

Active Sessions

Nunchaku Text to Image (Low VRAM)

Description

Discussion

Loading...