---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- output:
    url: images/d322c029-2dcc-48b2-9c5c-545352608f9c.png
  text: masterpiece, high_quality, highres, beauty,
  parameters:
    negative_prompt: worst quality, bad quality
- output:
    url: images/84177aa9-3f0c-4d67-9cdd-9893f7862c87.png
  text: masterpiece, high_quality, highres, beauty,
  parameters:
    negative_prompt: worst quality, bad quality
- output:
    url: images/e558b14d-6829-451c-a579-e182c69fade2.png
  text: masterpiece, high_quality, highres, beauty,
  parameters:
    negative_prompt: worst quality, bad quality
base_model: OnomaAIResearch/Illustrious-xl-early-release-v0
instance_prompt: None
license: cdla-permissive-2.0
---
# projectR_1

<Gallery />

## Model description 

![19402ba4-571b-4f64-a530-b8f699624563](https:&#x2F;&#x2F;cdn-uploads.huggingface.co&#x2F;production&#x2F;uploads&#x2F;663e3e78c719b0b8a386f24e&#x2F;h4K_dgU4ygfivpKEAEPAn.png)

Project:R —~
Not guided by direction.
But by curiosity.
A compass that doesn’t point north —~
but beyond the known.

━━━━━━━━━━━━━━━━━━━━

🧵 What makes this model special?

Honestly —~
that’s something you’ll discover best yourself.

Not by searching for answers.
But by following your imagination.

Your ideas.
Your stories.

━━━━━━━━━━━━━━━━━━━━

What I can tell you…

Project:R is the model that draws
what AI art and merging feels like to me.

Not just images.
But moments.

Moments —~ in between

Between styles.
Between worlds.
Between ideas.

━━━━━━━━━━━━━━━━━━━━

It’s not about mixing.

Merging, for me, is about coexistence.

Letting opposites exist —~
without forcing them to become one.

━━━━━━━━━━━━━━━━━━━━

A space where…

realism and stylization touch
control and randomness coexist
form and abstraction overlap
clarity and ambiguity live side by side

━━━━━━━━━━━━━━━━━━━━

The characters are not fully grounded.
But not fully lost either.

The worlds are not complete.
But not empty.

Everything exists —~
but not isolated.

━━━━━━━━━━━━━━━━━━━━

Project:R is not about control.

It’s about exploration.

━━━━━━━━━━━━━━━━━━━━

So what can you create with it?

That depends.

Not just on the model.

On you.

━━━━━━━━━━━━━━━━━━━━

You don’t need the perfect prompt.
You don’t need a fixed idea.
You don’t need to know where you’re going.

━━━━━━━━━━━━━━━━━━━━

Just start.

And see what happens.

━━━━━━━━━━━━━━━━━━━━

Project:R

Not an answer.
A space.

with ♥ Reij

🧵 Some of the threads woven into this model:

– isofocus —~ a balance between sharp presence and soft surroundings
– dynamic camera and movement —~ always in transition, never static
– duality in style —~ painted and rendered, existing side by side
– duality in focus —~ clarity and diffusion, not opposing but coexisting

🧵 Some of the settings I found myself returning to:

– DPM++ 2M Karras —~ a stable path through uncertainty
– 30–32 steps —~ enough detail to emerge, without over-defining
– CFG around 6–7 —~ guided, but not constrained

– Hires fix at ~1.5x —~ expanding the space, not sharpening control
– denoising around 0.3 —~ preserving the inbetween, not collapsing it

– format (~832×1216 &#x2F; ~832×1536) —~ reinforcing movement and flow

## Source

https://civitai.com/models/2495637/projectr?modelVersionId=2805426https://civitai.com/models/2495637/projectr?modelVersionId=2805426

## Credit

https://civitai.com/user/reijlita

## Trigger words

You should use `None` to trigger the image generation.


## Download model


[Download](/LyliaEngine/projectR_1/tree/main) them in the Files & versions tab.