SDXL Finetunes Img2img

Unleash creativity with SDXL Finetunes Img2img. Powered by SDXL and fine-tuning, it offers advanced image-to-image generation.

Platform: Replicate

Image-to-Image GenerationSDXL Fine-TuningLoRA ControlCreative Enhancement

52 runs

L40S

License Check Required

🚀Function Overview

Transforms input images based on text prompts while offering granular control over generation parameters like denoising, LoRA strengths, and safety filters.

Key Features

Detailed text prompt conditioning
Negative prompt filtering
Adjustable LoRA model strengths for style control
ControlNet and IP-Adapter configuration
Customizable image dimensions and output quality
Safety filter for NSFW content

Use Cases

•Creative image enhancement and restyling
•Product photo refinement
•Artistic transformation of existing images
•Detail-oriented photo manipulation

⚙️Input Parameters

input_image

string

Input image for processing. If not provided, the default image will be used.

prompt

string

High quality modifiers to add to prompt

negative_prompt

string

Things you do not want to see in your image

checkpoint_name

string

Checkpoint model to use

steps

integer

Number of inference steps

cfg

number

Guidance scale

denoise

number

Denoising strength

sampler_name

string

Sampling algorithm to use

scheduler

string

Scheduler for the sampler

lora1_strength_model

number

LoRA 1 (product marketing) model strength

lora1_strength_clip

number

LoRA 1 (product marketing) clip strength

lora2_strength_model

number

LoRA 2 (add-detail-xl) model strength

lora2_strength_clip

number

LoRA 2 (add-detail-xl) clip strength

lora3_strength_model

number

LoRA 3 (detailed_notrigger) model strength

lora3_strength_clip

number

LoRA 3 (detailed_notrigger) clip strength

ipadapter_weight

number

IP-Adapter weight

ipadapter_start_at

number

IP-Adapter start at step percentage

ipadapter_end_at

number

IP-Adapter end at step percentage

controlnet_strength

number

ControlNet strength

controlnet_start_percent

number

ControlNet start percentage

controlnet_end_percent

number

ControlNet end percentage

safety_filter

boolean

Enable Replicate safety filter to detect and filter NSFW content (can't be disabled in web UI)

width

integer

Width of the image

height

integer

Height of the image

output_format

string

Format of the output images

output_quality

integer

Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality.

seed

integer

Set a seed for reproducibility. Random by default.

💡Usage Examples

Example 1

Input Parameters

Output Results

Quick Actions

Use NowView Documentation

Technical Specifications

Hardware Type: L40S
Run Count: 52
Commercial Use: Unknown/Restricted
Platform: Replicate

Related Keywords

Image-to-Image GenerationSDXL Fine-TuningLoRA ControlCreative EnhancementDetailed Text Prompt ConditioningCustomizable Image DimensionsProduct Photo RefinementArtistic Transformation

Related Models

Cartoonize Image with FLUX.1 Kontext

Turn your image into a cartoon with FLUX.1 Kontext [pro]

Mirage Ghibli Style Image Generator

Ghiblify any image, 10x cheaper/faster than GPT 4o

Luma Reframe Image

Change the aspect ratio of any photo using AI (not cropping)