Virtual product
{{ variable.name }}
Select products to buy together. Choose each variant and quantity.
Quantities are the totals for this purchase. Saving combines the items into one bundle and recalculates the discount.
{{ bundleConfig.name }} Main product
Variant and quantity follow your selection above · × {{ quantity }}
{{ bundleStockLabel(product) }}
{{ bundleStockError(product, quantity) }}
{{ bundleLimitError(product, quantity) || bundleLimitLabel(product) }}
{{ item.name }}
{{ bundleStockLabel(bundleSku(item)) }}
{{ bundleStockError(bundleSku(item), item.quantity) }}
Check the selected variants, quantities and stock.
{{ (item.checked && bundleLimitError(bundleSku(item), item.quantity)) || bundleLimitLabel(bundleSku(item)) }}
The main product is excluded. Multiple units of the same product count as one style.
All complementary products have been deselected. Saving will change this to a separate purchase of the main product, without the bundle discount.
{{ bundleSelectedLabel }} Highest eligible discount applied
{{ bundleSelectionError || bundleError }}
Unable to calculate. Resolve the message above first.
Based on your current selection. Final discounts are confirmed at checkout.
Quantity: {{ quantity * bundleSetQuantity }}
Sets: {{ bundleSetQuantity }}
Customize the main product in your cart or at checkout. Customization fees are charged separately.
{{ bundleError }}
Loading…
A comprehensive step-by-step guide to training custom LoRA models for Stable Diffusion 1.5 — the most mature, resource-friendly, and community-supported AI image generation model. Perfect for beginners entering the world of "alchemy" (model training) with minimal hardware requirements.
Product Type: Digital Tutorial (HTML + Markdown)
Why SD1.5 for Training?
- GPU-friendly: Works on 6GB+ VRAM (most accessible entry point)
- Massive ecosystem: Largest community of plugins, checkpoints, and LoRAs
- Fast training: Complete a LoRA in 20-40 minutes on mid-range GPUs
- Proven stability: Years of community refinement and documentation
Tutorial Structure (6 Chapters):
【Chapter 1: Environment & Hardware】
- GPU requirements: NVIDIA 6GB+ VRAM (8GB recommended)
- Kohya_ss GUI installation with path requirements
- Storage: 20GB+ SSD space for models and outputs
【Chapter 2: Dataset Preparation】
- Image specifications: 15-40 images, 512x512 resolution
- WD14 Tagger for automatic captioning
- Trigger word setup and tag cleaning strategies
- Folder structure: Strict Kohya naming convention (Repeats_ConceptName)
【Chapter 3: Configuration (YAML)】
- Complete parameter breakdown with explanations
- Network settings: Rank 32-128, Alpha recommendations
- Learning rate tuning: Unet vs Text Encoder rates
- Clip Skip settings: 2 for Anime, 1 for Realistic
【Chapter 4: Training Process】
- Step-by-step execution guide
- Loss value interpretation (0.08-0.15 normal range)
- Time estimates: ~20-40 mins for 3000 steps on RTX 3060
【Chapter 5: Model Pruning】
- Removing optimizer states for standard LoRA size
- Expected file size: ~36MB for Rank 32
【Chapter 6: Testing & Application】
- XY Plot testing methodology
- Weight recommendations: 0.6-0.8 (1.0 often burns images)
- Trigger word usage in prompts
Bonus: FAQ & Troubleshooting
- OOM solutions for low VRAM users
- Grey/oversaturated image fixes
- Character resemblance issues
- Face corruption remedies
Technical Requirements:
- NVIDIA GPU with 6GB+ VRAM (8GB+ recommended)
- Windows with Kohya_ss GUI
- 20GB+ free storage space
Package Contents:
- SD1.5 Legacy LoRA Training Complete Tutorial.html (formatted tutorial)
- SD1.5 Legacy LoRA Training Complete Tutorial.md (markdown source)
Who This Is For:
- Complete beginners starting their first LoRA training
- Artists wanting to train anime character or realistic style LoRAs
- Users with limited GPU resources (6-8GB VRAM)
- Anyone seeking the most accessible entry point to AI model training