Deploy Dedicated GPU server to run AI models

Deploy Model
Skip to main content
UrangDiffusion v3.1 - v1.2 thumbnail

UrangDiffusion V3.1 - V1.2

by ModelsLab

Training Configurations

Finetuned from: Animagine XL 3.1

Pretraining:

  • Dataset size: ~23,500 images

  • GPU: 1xA100 80GB

  • Optimizer: AdaFactor

  • Unet Learning Rate: 3.75e-6

  • Text Encoder Learning Rate: 1.875e-6

  • Batch Size: 48

  • Gradient Accumulation: 1

  • Warmup steps: 100 steps

  • Min SNR: 5

  • Epoch: 10 (epoch 9 is used)

Finetuning:

  • Dataset size: ~6,800 images

  • GPU: 1xA100 80GB

  • Optimizer: AdaFactor

  • Unet Learning Rate: 2e-6

  • Text Encoder Learning Rate: - (Train TE set to False)

  • Batch Size: 48

  • Gradient Accumulation: 1

  • Warmup steps: 5%

  • Min SNR: 5

  • Epoch: 10

  • Noise Offset: 0.0357

Wuthering Waves, Zenless Zone Zero and hololiveEN -Justice- have been added to the model.

urangdiffusion-v3-1-v1-2
Open Source ModelUnlimited UsageLLMs.txt
API PlaygroundAPI Documentation

Input

Select models...

Per image generation will cost 0.0047$
For premium plan image generation will cost 0.00$ i.e Free.

Output

Idle

Unknown content type

Related Models

Discover similar models you might be interested in

View all Models
Sinfully Stylish (dramatic lighting) - v1.0 forSDXL
Popular
ModelsLab

Sinfully Stylish (dramatic lighting) - v1.0 forSDXL

Open Source Model
Anime Illust Diffusion XL - v0.7
Popular
ModelsLab

Anime Illust Diffusion XL - v0.7

Open Source Model
Real Dream SDXL
ModelsLab

Real Dream SDXL

Open Source Model
AingDiffusion XL
Popular
ModelsLab

AingDiffusion XL

Open Source Model
DC Villains Series: Catwoman - Selina Kyle v1.0
ModelsLab

DC Villains Series: Catwoman - Selina Kyle v1.0

Open Source Model
JJ's Interior Space - Kids room - XL v1.0
Popular
ModelsLab

JJ's Interior Space - Kids room - XL v1.0

Open Source Model
Better Picture, More Details LoRA - v5
Popular
ModelsLab

Better Picture, More Details LoRA - v5

Open Source Model
[Illustrious / SDXL] Super robot diffusion XL (Gundam, EVA, ARMORED CORE, BATTLE TECH like mecha lor
Popular
ModelsLab

[Illustrious / SDXL] Super robot diffusion XL (Gundam, EVA, ARMORED CORE, BATTLE TECH like mecha lor

Open Source Model
Alma Photographic XL - v1
ModelsLab

Alma Photographic XL - v1

Open Source Model
C4t3c0l0r - V1
ModelsLab

C4t3c0l0r - V1

Open Source Model
Rotated View (Dutch Angle) Cinematic Style XL + F1D - Dutch Angle XL v2.0
Popular
ModelsLab

Rotated View (Dutch Angle) Cinematic Style XL + F1D - Dutch Angle XL v2.0

Open Source Model
Object Showcase Collection of 60 Items [XL] - 03 Tablet Showcase
Popular
ModelsLab

Object Showcase Collection of 60 Items [XL] - 03 Tablet Showcase

Open Source Model
Cinematic Volumetric (God Rays) Lighting Style XL + F1D + Illu + Pony - Volumetric xl v2
Popular
ModelsLab

Cinematic Volumetric (God Rays) Lighting Style XL + F1D + Illu + Pony - Volumetric xl v2

Open Source Model
The Deep Dark Dungeon-Pony + SDXL - v1-SDXL
Popular
ModelsLab

The Deep Dark Dungeon-Pony + SDXL - v1-SDXL

Open Source Model
blue_pencil-XL - v0.1.0
Popular
ModelsLab

blue_pencil-XL - v0.1.0

Open Source Model
Tst
ModelsLab

Tst

Open Source Model
dawn_(pokemon)_noobxl_eps_1.0 - v1.0
ModelsLab

dawn_(pokemon)_noobxl_eps_1.0 - v1.0

Open Source Model
Sevenof9_photo_art_fantasy2 - SDXL_myown_v1_Sevenof9
ModelsLab

Sevenof9_photo_art_fantasy2 - SDXL_myown_v1_Sevenof9

Open Source Model

UrangDiffusion V3.1 - V1.2 Readme

Training Configurations

Finetuned from: Animagine XL 3.1

Pretraining:

  • Dataset size: ~23,500 images

  • GPU: 1xA100 80GB

  • Optimizer: AdaFactor

  • Unet Learning Rate: 3.75e-6

  • Text Encoder Learning Rate: 1.875e-6

  • Batch Size: 48

  • Gradient Accumulation: 1

  • Warmup steps: 100 steps

  • Min SNR: 5

  • Epoch: 10 (epoch 9 is used)

Finetuning:

  • Dataset size: ~6,800 images

  • GPU: 1xA100 80GB

  • Optimizer: AdaFactor

  • Unet Learning Rate: 2e-6

  • Text Encoder Learning Rate: - (Train TE set to False)

  • Batch Size: 48

  • Gradient Accumulation: 1

  • Warmup steps: 5%

  • Min SNR: 5

  • Epoch: 10

  • Noise Offset: 0.0357

Wuthering Waves, Zenless Zone Zero and hololiveEN -Justice- have been added to the model.