Giving 200$ to anyone making a runpod for the community.

[Anashel-RPG]

Greetings!

I've spent the entire day trying to get a Runpod instance working, but I keep encountering the error mentioned in this thread: #703.

I'm willing to pay someone to help set up a proper Runpod so the community can use it effectively. I need Syncthing and SimpleTuner to work correctly for the recommended instance type.

Please provide simple instructions for uploading an image set and getting it up and running. SimpleTuner is fantastic, but I hope we can make it work for more people without losing an entire day. :)

Tag me here if you want to take the challenge.

[Anashel-RPG]

Manage to made it work with this config. I will make a runpod template for everyone. :)

# SimpleTuner Configuration File
## Version 002 for RunPod

# Model Configuration
#######################
export MODEL_TYPE='lora'
export MODEL_NAME="black-forest-labs/FLUX.1-dev"
export FLUX=true

# Training Parameters
#######################
export LEARNING_RATE=1e-6
export MAX_NUM_STEPS=2000
export NUM_EPOCHS=0
export TRAIN_BATCH_SIZE=4
export GRADIENT_ACCUMULATION_STEPS=2
export VAE_BATCH_SIZE=4

# Checkpointing and Validation
###############################
export CHECKPOINTING_STEPS=500
export CHECKPOINTING_LIMIT=2
export RESUME_CHECKPOINT="latest"
export VALIDATION_STEPS=500
export VALIDATION_NUM_INFERENCE_STEPS=30
export VALIDATION_SEED=42

# Model Specific Settings
##########################
export STABLE_DIFFUSION_3=false
export PIXART_SIGMA=false
export STABLE_DIFFUSION_LEGACY=false
export KOLORS=false
export CONTROLNET=false
export USE_DORA=false

# Output and Logging
#####################
export OUTPUT_DIR="output/models"
export DEBUG_EXTRA_ARGS="--report_to=wandb"
export TRACKER_PROJECT_NAME="${MODEL_TYPE}-training"
export TRACKER_RUN_NAME="simpletuner-sdxl"

# Hugging Face Hub Integration
###############################
export PUSH_TO_HUB="false"
export PUSH_CHECKPOINTS="false"
export HUB_MODEL_NAME=$TRACKER_PROJECT_NAME

# Image Processing
###################
export RESOLUTION=1024
export RESOLUTION_TYPE="pixel"
export MINIMUM_RESOLUTION=$RESOLUTION

# Prompts and Captions
#######################
export VALIDATION_PROMPT="ethnographic photography of teddy bear at a picnic"
export VALIDATION_GUIDANCE=7.5
export VALIDATION_GUIDANCE_RESCALE=0.0
export VALIDATION_NEGATIVE_PROMPT="blurry, cropped, ugly"
export CAPTION_DROPOUT_PROBABILITY=0.1

# Learning Rate Schedule
#########################
export LR_SCHEDULE="polynomial"
export LR_WARMUP_STEPS=1000

# Performance Optimizations
############################
export MAX_WORKERS=32
export READ_BATCH_SIZE=25
export WRITE_BATCH_SIZE=64
export IMAGE_PROCESSING_BATCH_SIZE=35
export AWS_MAX_POOL_CONNECTIONS=128
export TORCH_NUM_THREADS=8

# Advanced Training Options
############################
export TRAINING_SCHEDULER_TIMESTEP_SPACING="trailing"
export INFERENCE_SCHEDULER_TIMESTEP_SPACING="trailing"
export MIN_SNR_GAMMA=5
export USE_XFORMERS=false
export USE_GRADIENT_CHECKPOINTING=true
export ALLOW_TF32=true
export OPTIMIZER="adamw_bf16"
export USE_EMA=false
export EMA_DECAY=0.999

# Additional Training Arguments
################################
export TRAINER_EXTRA_ARGS="--lora_rank=4"

# Reproducibility
##################
export TRAINING_SEED=42

# Precision Settings
#####################
export MIXED_PRECISION="bf16"
export PURE_BF16=true

# Distributed Training
#######################
export TRAINING_NUM_PROCESSES=1
export TRAINING_NUM_MACHINES=1
export ACCELERATE_EXTRA_ARGS=""

# Compilation Options
######################
export TRAINING_DYNAMO_BACKEND='no'
export TOKENIZERS_PARALLELISM=false

# Data Management
##################
export DELETE_ERRORED_IMAGES=0
export DELETE_SMALL_IMAGES=0
export DATALOADER_CONFIG="config/multidatabackend.json"
export METADATA_UPDATE_INTERVAL=65