---
title: PlantyMiji
canonical_url: "https://www.modelscope.ai/models/Whyxeo/PlantyMiji"
md_url: "https://www.modelscope.ai/models/Whyxeo/PlantyMiji.md"
repository: Whyxeo/PlantyMiji
last_updated: 2026-07-12
license: other
pipeline_tag: text-to-image-synthesis
tasks:
  - text-to-image-synthesis
base_model:
  - Whyxeo/illustrious_Whyx
base_model_relation: merge
parameters: 6.9B
tensor_type:
  - F16
library_name:
  - safetensors
  - other
frameworks:
  - other
language:
  - ru
  - be
  - en
downloads: 63
stars: 0
tags:
  - ckpt
  - safetensors
  - model
  - styled
  - miji
  - illustrious
  - v10
---

# PlantyMiji

> PlantyMiji - An open-source model by Whyxeo on ModelScope. NeoMiji — v10.0

Whyxeo/PlantyMiji is a 6.9B-parameter text-to-image-synthesis model on ModelScope. licensed under other. derived from Whyxeo/illustrious_Whyx.

- **Repository**: Whyxeo/PlantyMiji
- **License**: other
- **Tasks**: text-to-image-synthesis
- **Parameters**: 6.9B
- **Base model**: Whyxeo/illustrious_Whyx
- **Tags**: ckpt, safetensors, model, styled, miji, illustrious, v10
- **Downloads**: 63
- **Stars**: 0
- **Last updated**: 2026-07-12

Source: https://www.modelscope.ai/models/Whyxeo/PlantyMiji

---

<div align="center">

# NeoMiji v10.0

### Illustrious-based · Final release of the NeoMiji Illustrious line

![Base Model](https://img.shields.io/badge/Base%20Model-Plant%20Milk%20Walnut-8A2BE2?style=flat-square)
![Architecture](https://img.shields.io/badge/Architecture-Illustrious-orange?style=flat-square)
![Version](https://img.shields.io/badge/Version-10.0-blue?style=flat-square)
![Status](https://img.shields.io/badge/Status-Final%20Release-red?style=flat-square)

</div>

---

## Overview

| | |
|---|---|
| **Base model** | Plant Milk Walnut |
| **Integration method** | Fine-Tune LoRA |
| **Dataset** | Custom dataset (originally built for NeoMiji v2.0, retrained) |
| **Line** | NeoMiji · Illustrious branch (final entry) |
| **Recommended style-LoRA strength** | 0.6 – 0.8 |

NeoMiji v10.0 is built on **Plant Milk Walnut** as the base checkpoint. The custom dataset originally developed for **NeoMiji v2.0** was reused, retrained, and cleanly integrated back into the model through **Fine-Tune LoRA**.

Compared to earlier entries in the line, this version brings a noticeable step up in:

- 🎨 **Image quality** — cleaner output with fewer artifacts
- 🧠 **Prompt understanding** — more accurate interpretation of complex prompts
- 🧍 **Anatomy & proportions** — more consistent, coherent structure
- 🔍 **Detail level** — sharper fine detail across the image

Compatibility with **style LoRAs** has been preserved from earlier versions — but use them carefully. A reduced strength of **0.6–0.8** is recommended so the style LoRA enhances rather than overrides the base model's own quality.

---

## About This Release

This model is a separate branch within the NeoMiji lineage, and marks the **last of two models built on the Illustrious architecture** — no further Illustrious-based continuations are currently planned. The Illustrious base has effectively plateaued in terms of what can be extracted from it, and this release represents that ceiling.

Future development is now focused on the **Anima** architecture, which handles both stylistic range and anatomical coherence noticeably better going forward.

> **v10.0 is considered the refined, improved successor to NeoMiji v2.0** — same dataset lineage, rebuilt from the ground up on a stronger base and integration method.

---

## Access & License

Please read the license carefully **before downloading or using this model**. It is available in two languages within this repository:

| Language | File |
|---|---|
| 🇬🇧 English | `WhyxLicense_LICENSE_EN.md` |
| 🇷🇺 Russian | `WhyxLicense_LICENSE_RU.md` |

By downloading or using this model, you agree to comply with the **Whyx License** terms in full.

---

## Recommended Generation Settings

### Core Parameters

| Parameter | Value |
|---|---|
| Sampling Method | Euler a |
| Schedule Type | SGM Uniform |
| Sampling Steps | 30 |
| Width | 512 |
| Height | 768 |
| CFG Scale | 3.5 |
| Seed | -1 (random) |
| Batch Count | 1 |
| Batch Size | 1 |

### Hires. Fix

| Parameter | Value |
|---|---|
| Enabled | ✔️ |
| Resolution | 512×768 → 1024×1536 |
| Upscaler | 4x_foolhardy_Remacri |
| Upscale by | 2 |
| Hires Steps | 15 |
| Denoising Strength | 0.45 |
| Hires CFG Scale | 3.5 |
| Resize width to | 0 |
| Resize height to | 0 |

### ADetailer

| Parameter | Value |
|---|---|
| Enabled | ✔️ |
| Version | Neo |

---

<div align="center">

*NeoMiji — Illustrious line, final release · Development continues on Anima*

</div>