Flux Lora Training Model - Open-source for Text-to-Image and Image-to-Image Generation, Supports High-resolution Image Generation

Flux Lora Training

Developed by hok00i3

This is a PEFT LoRA model based on the FLUX.1-dev model, focusing on text-to-image and image-to-image tasks, supporting high-resolution image generation.

Image Generation Open Source License:Other #Photorealistic #Anime Character Generation #High-Resolution Images

Downloads 39

Release Time : 1/10/2025

Model Overview

This model is a standard PEFT LoRA derived from the FLUX.1-dev base model, primarily used for generating high-quality photorealistic images, especially excelling in creating detailed images of Japanese women in their early twenties.

Model Features

High-Resolution Image Generation

Supports 4K and 8K high-resolution image generation with photorealistic quality.

Fine Control

Allows precise control over image details such as clothing, expressions, and scenes through detailed prompts.

LoRA Fine-Tuning

Utilizes LoRA technology for fine-tuning, enhancing specific style generation capabilities while preserving the base model's performance.

Model Capabilities

Text-to-Image

Image-to-Image

High-Resolution Image Generation

Photorealistic Image Generation

Use Cases

Art Creation

Anime-Style Character Generation

Generates detailed anime-style character images, allowing control over clothing, expressions, and other details.

Example images showcase a female character in a sailor uniform in a classroom setting.

Concept Design

Character Design

Quickly generates character concept art for use in game or animation pre-production.

🚀 flux-lora-training

This project is a standard PEFT LoRA derived from black-forest-labs/FLUX.1-dev, which is mainly used for text - to - image and image - to - image tasks.

Metadata

Property	Details
License	other
Base Model	black-forest-labs/FLUX.1-dev
Pipeline Tag	text-to-image
Inference	true

Widget Examples

Unconditional (blank prompt):
- Negative Prompt: ''
- Output: image
Specific Prompt:
- Prompt: '(ultra detailed,photo,photograph,best quality,high resolution,4k,8k,photorealistic,Japanese early twenties,(slim) and curvy body,waist,detailed beautiful eyes,super detailed eyes and skins,very beautiful woman:3.0), (Wearing tight short sleeves white (one piece) sailor uniform, blue collar, red neckerchief, dark blue pleated dress, standing in empty classroom,sweet smiling:2.0), long straight hair, standing in empty classroom,'
- Negative Prompt: ''
- Output: image

🚀 Quick Start

Validation Prompt

The main validation prompt used during training was:

(ultra detailed,photo,photograph,best quality,high resolution,4k,8k,photorealistic,Japanese early twenties,(slim) and curvy body,waist,detailed beautiful eyes,super detailed eyes and skins,very beautiful woman:3.0), (Wearing tight short sleeves white (one piece) sailor uniform, blue collar, red neckerchief, dark blue pleated dress, standing in empty classroom,sweet smiling:2.0), long straight hair, standing in empty classroom,

Validation Settings

CFG: 3.5
CFG Rescale: 0.0
Steps: 20
Sampler: FlowMatchEulerDiscreteScheduler
Seed: 42
Resolution: 1024x1024
Skip - layer guidance:

Note: The validation settings are not necessarily the same as the training settings.

You can find some example images in the following gallery:

The text encoder was not trained. You may reuse the base model text encoder for inference.

📦 Training Settings

Training epochs: 92
Training steps: 2400
Learning rate: 0.0001
- Learning rate schedule: constant_with_warmup
- Warmup steps: 100
Max grad value: 1.0
Effective batch size: 1
- Micro - batch size: 1
- Gradient accumulation steps: 1
- Number of GPUs: 1
Gradient checkpointing: True
Prediction type: flow - matching (extra parameters=['flow_schedule_auto_shift', 'shift=0.0', 'flux_guidance_mode=constant', 'flux_guidance_value=1.0', 'flow_matching_loss=compatible', 'flux_lora_target=all+ffs'])
Optimizer: adamw_bf16
Trainable parameter precision: Pure BF16
Base model precision: no_change
Caption dropout probability: 0.0%
LoRA Rank: 16
LoRA Alpha: None
LoRA Dropout: 0.1
LoRA initialisation style: default

📚 Datasets

cissy

Repeats: 0
Total number of images: 26
Total number of aspect buckets: 1
Resolution: 1.0 megapixels
Cropped: true
Crop style: center
Crop aspect: square
Used for regularisation data: No

💻 Usage Examples

Basic Usage

import torch
from diffusers import DiffusionPipeline

model_id = 'black-forest-labs/FLUX.1-dev'
adapter_id = 'hok00i3/flux-lora-training'
pipeline = DiffusionPipeline.from_pretrained(model_id, torch_dtype=torch.bfloat16) # loading directly in bf16
pipeline.load_lora_weights(adapter_id)

prompt = "(ultra detailed,photo,photograph,best quality,high resolution,4k,8k,photorealistic,Japanese early twenties,(slim) and curvy body,waist,detailed beautiful eyes,super detailed eyes and skins,very beautiful woman:3.0), (Wearing tight short sleeves white (one piece) sailor uniform, blue collar, red neckerchief, dark blue pleated dress, standing in empty classroom,sweet smiling:2.0), long straight hair, standing in empty classroom,"


## Optional: quantise the model to save on vram.
## Note: The model was not quantised during training, so it is not necessary to quantise it during inference time.
#from optimum.quanto import quantize, freeze, qint8
#quantize(pipeline.transformer, weights=qint8)
#freeze(pipeline.transformer)
    
pipeline.to('cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu') # the pipeline is already in its target precision level
model_output = pipeline(
    prompt=prompt,
    num_inference_steps=20,
    generator=torch.Generator(device='cuda' if torch.cuda.is_available() else 'mps' if torch.backends.mps.is_available() else 'cpu').manual_seed(42),
    width=1024,
    height=1024,
    guidance_scale=3.5,
).images[0]

model_output.save("output.png", format="PNG")

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご