ProjectIL v3 Image Generation

Discover the power of ProjectIL v3 Image Generation. Create stunning visuals with advanced text-to-image synthesis and intricate refinement tools.

Platform: Replicate

Text-to-Image GenerationAI Art CreationImage RefinementPrompt Engineering

2.8k runs

L40S

License Check Required

🚀Function Overview

Generates images from text descriptions with extensive controls for quality, composition, and refinement using parameters like CFG scaling, CLIP skip, and detail enhancement tools.

Key Features

Text-to-image synthesis with Compel prompt weighting
Negative prompting for content exclusion
Multi-object detail refinement (face/hand/person)
Image upscaling and refiner pipeline support
Parameter controls for dimensions, sampler, steps, and seeds

Use Cases

•Generating digital artwork from descriptive prompts
•Creating character designs for games/media
•Prototyping visual concepts with style controls
•Enhancing image details through post-processing

⚙️Input Parameters

model

string

The model to use

vae

string

The VAE to use

prompt_conjunction

boolean

Force Compel conjunction syntax on prompt, separating linebreaks into different items on the Compel conjunction function

prompt

string

The prompt, uses Compel weighting syntax

negative_prompt

string

The negative prompt (for things you don't want), uses Compel weighting syntax

negative_prompt_conjunction

boolean

Force Compel conjunction syntax on prompt, separating linebreaks into different items on the Compel conjunction function

cfg_scale

number

CFG scale defines how much attention the model pays to the prompt when generating, set to 1 to disable

guidance_rescale

number

Amount to rescale CFG generated noise to avoid overexposed images, set to 0 or 1 to disable

pag_scale

number

PAG scale is similar to CFG but improves the result; set to 0 to disable

clip_skip

integer

How many CLIP layers to skip; 1 means no skip

width

integer

The width of the image

height

integer

The height of the image

upscale

string

Upscale factor

refiner

boolean

Enable image refiner using model

refiner_prompt

string

The prompt of refiner (leave empty for default correct hands and details)

refiner_strength

number

Refiner strength (0.0 to 1.0)

prepend_preprompt

boolean

Prepend preprompt (Prompt: "masterpiece, best quality, absurdres," | Negative prompt: "low quality, bad quality, watermark,")

scheduler

string

The scheduler to use

steps

integer

Number of steps when generating

seed

integer

The seed for generation; set to -1 for random seed

adetailer_face

boolean

Use ADetailer face_yolov9c.pt

adetailer_face_prompt

string

ADetailer face prompt (if blank, will use the image prompt)

adetailer_face_negative_prompt

string

ADetailer face negative prompt (if blank, will use: blurry, deformed face, extra features, unrealistic skin texture, oversmoothed, low resolution, artifacts)

adetailer_hand

boolean

Use ADetailer hand_yolov9c.pt

adetailer_hand_prompt

string

ADetailer hand prompt (if blank, will use: detailed hands, realistic anatomy, well-defined fingers, natural pose, soft shadows, high resolution, intricate details)

adetailer_hand_negative_prompt

string

ADetailer hand negative prompt (if blank, will use: extra fingers, fused or missing fingers, blurry, distorted, low detail, poorly drawn, unnatural proportions)

adetailer_person

boolean

Use ADetailer person_yolov8m-seg.pt

adetailer_person_prompt

string

ADetailer person prompt (if blank, will use: best quality, detailed, realistic shadows)

adetailer_person_negative_prompt

string

ADetailer person negative prompt (if blank, will use: deformed, unrealistic proportions, extra limbs, blurred, low quality, distorted pose, exaggerated features)

💡Usage Examples

Example 1

Input Parameters

{
  "vae": "default",
  "seed": -1,
  "model": "ProjectIL-v3",
  "steps": 30,
  "width": 1024,
  "height": 1024,
  "prompt": "street, 1girl, dark-purple short hair, purple eyes, medium breasts, cleavage, casual clothes, smile",
  "refiner": false,
  "upscale": "Original",
  "cfg_scale": 5,
  "clip_skip": 0,
  "pag_scale": 0,
  "scheduler": "DPM++ 2M SDE Karras",
  "adetailer_face": false,
  "adetailer_hand": false,
  "refiner_prompt": "",
  "negative_prompt": "nsfw, naked",
  "adetailer_person": false,
  "guidance_rescale": 0,
  "refiner_strength": 0.4,
  "prepend_preprompt": true,
  "prompt_conjunction": false,
  "adetailer_face_prompt": "",
  "adetailer_hand_prompt": "",
  "adetailer_person_prompt": "",
  "negative_prompt_conjunction": false,
  "adetailer_face_negative_prompt": "",
  "adetailer_hand_negative_prompt": "",
  "adetailer_person_negative_prompt": ""
}

Output Results

https://replicate.delivery/xezq/j8fpFX1URd3yBKW33womVbef3oRfGnF9S1w3QC9t1zsy8U2RB/0.png

Quick Actions

Use NowView Documentation

Technical Specifications

Hardware Type: L40S
Run Count: 2.8k
Commercial Use: Unknown/Restricted
Platform: Replicate

Related Keywords

Text-to-Image GenerationAI Art CreationImage RefinementPrompt EngineeringCharacter DesignVisual Concept PrototypingDigital ArtworkImage Upscaling

Related Models

f-beta.0 Image Generation

A model for generating and editing images using diffusion-based techniques with LoRA support.

Regina Image Generation and Editing Model

A model for generating and editing images based on text prompts and input images.

Elena AI Image Generator

A model for generating and editing images through text prompts, input images, masks, and LoRA integrations.