README.md

---
language:
- en
license: other
license_name: fair-ai-public-license-1.0-sd
license_link: https://freedevproject.org/faipl-1.0-sd/
base_model:
- Laxhar/noobai-XL-Vpred-1.0
- MidnightRunner/MIDNIGHT_NAI-XL_vPredV1
- RedRayz/hikari_noob_v-pred_1.2.1
- motimalu/kirazuri-lazuli-noobai-xl-vpred-2.0
- motimalu/kirazuri-lazuli-noobai-xl-vpred
- John6666/wai-shuffle-noob-vpred20-sdxl
- cyberdelia/CyberRealisticXL
pipeline_tag: text-to-image
library_name: diffusers
tags:
- safetensors
- diffusers
- stable-diffusion
- stable-diffusion-xl
- merge
---

<style>
  .container {
    display: flex;
    gap: 2rem;
    align-items: flex-start;
  }
  .image-wrapper,
  .details {
    flex: 1 1 50%;
  }
  .image-wrapper {
    display: flex;
    justify-content: center;
    align-items: center;
  }
  .model-image {
    width: 100%;
    max-width: 500px;
    min-width: 300px;
    border-radius: 8px;
    box-shadow: 0 8px 20px rgba(0, 0, 0, 0.3);
  }
  .details h1 {
    margin-top: 0;
    font-size: 2rem;
    color: #fff;
  }
  .details .description {
    margin: 0.5rem 0 1.5rem;
    line-height: 1.5;
    color: #ccc;
  }
  .features-list {
    list-style: none;
    padding: 0;
    margin: 0 0 1.5rem;
  }
  .features-list li {
    display: flex;
    align-items: center;
    margin-bottom: 0.5rem;
    font-size: 1rem;
    color: #fff;
  }
  .features-list .icon {
    margin-right: 0.6rem;
    font-size: 1.2rem;
  }
  .badge-list {
    list-style: none;
    padding: 0;
    margin: 0;
    display: flex;
    flex-wrap: wrap;
    row-gap: 0.25rem;
    column-gap: 0.5rem;
  }
  .badge-list li {
    margin: 0;
  }
  .badge-btn {
    display: inline-flex;
    align-items: center;
    height: 28px;
    padding: 0 10px;
    font-size: 14px;
    font-weight: 500;
    color: #fff;
    text-decoration: none;
    border-radius: 4px;
    white-space: nowrap;
    transition: opacity 0.2s, transform 0.1s;
  }
  .badge-btn .btn-icon {
    width: 16px;
    height: 16px;
    margin-right: 6px;
  }
  .badge-btn:hover {
    opacity: 0.85;
    transform: translateY(-1px);
  }
  .badge-btn.civitai {
    background-color: #2975fa8d;
  }
  .badge-btn.hf {
    background-color: #ff66008d;
  }

  @media (max-width: 700px) {
    .container {
      flex-direction: column;
      align-items: center;
    }
    .image-wrapper,
    .details {
      flex: unset;
      width: 100%;
    }
    .model-image {
      width: 90%;
      max-width: 500px;
      min-width: auto;
      margin-bottom: 1rem;
    }
  }
</style>

<div class="container">
  <div class="image-wrapper">
    <img
      src="https://huggingface.co/rityak/NAI-X-vpred/resolve/main/naix_00002_.png"
      alt="NAI-X Example"
      class="model-image"
    />
  </div>
  <div class="details">
    <h1>NAI‑X v2.0</h1>
    <p class="description">
      Refined and more stable version of <strong>NoobAI‑XL V‑Pred</strong>, delivering
      predictable, consistent results even with simple or dark prompts.
    </p>
    <ul class="features-list">
      <li><span class="icon">🛠️</span><strong>Improved prompt tolerance</strong></li>
      <li><span class="icon">🎨</span><strong>Enhanced color & lighting balance</strong></li>
      <li><span class="icon">📐</span><strong>Stable anatomy & layout</strong></li>
      <li><span class="icon">🔄</span><strong>Faithful to original V‑Pred</strong></li>
    </ul>
    <ul class="badge-list">
      <li>
        <a
          href="https://civitai.com/models/1804251/nai-x-v20-vanilla-like-noob"
          target="_blank"
          class="badge-btn civitai"
        >
          <img
            src="https://civitai.com/favicon-blue.ico"
            alt="CivitAI"
            class="btn-icon"
          />
          View on CivitAI
        </a>
      </li>
      <li>
        <a
          href="https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAIX-v20_vpred.safetensors?download=true"
          target="_blank"
          class="badge-btn hf"
        >
          <img
            src="https://huggingface.co/front/assets/huggingface_logo-noborder.svg"
            alt="Hugging Face"
            class="btn-icon"
          />
          Download NAI‑X v2.0
        </a>
      </li>
      <li>
        <a
          href="https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAI-X-soft-vpred.safetensors?download=true"
          target="_blank"
          class="badge-btn hf"
        >
          <img
            src="https://huggingface.co/front/assets/huggingface_logo-noborder.svg"
            alt="Hugging Face"
            class="btn-icon"
          />
          Download NAI‑X‑soft (v1.0‑soft)
        </a>
      </li>
      <li>
        <a
          href="https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAI-X-zero-vpred.safetensors?download=true"
          target="_blank"
          class="badge-btn hf"
        >
          <img
            src="https://huggingface.co/front/assets/huggingface_logo-noborder.svg"
            alt="Hugging Face"
            class="btn-icon"
          />
          Download NAI‑X‑zero (v1.0‑zero)
        </a>
      </li>
    </ul>
  </div>
</div>


---
## Model Variants

- **[NAI-X v2.0](https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAIX-v20_vpred.safetensors?download=true)** 
  The latest and most balanced version. Closely aligned with the original V-Pred base, but cleaner, more stable, and easier to use. Built using a simplified merging recipe for better prompt robustness and visual quality.

- **[NAI-X-soft (v1.0-soft)](https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAI-X-soft-vpred.safetensors?download=true)**  
  A light refinement of NoobAI-XL V-Pred. Focuses on composition, anatomy, and color improvements while keeping all original tokens and tag behavior intact.

- **[NAI-X-zero (v1.0-zero)](https://huggingface.co/rityak/NAI-X-vpred/resolve/main/NAI-X-zero-vpred.safetensors?download=true)**  
  A stronger, more expressive variant. Adds LoRA DPO and deeper changes in rendering logic. Sharper detail and stronger prompt influence, especially on faces and backgrounds.

---
## Recommended Settings

To get the best results from **NAI-X**, use the parameters below. These match the internal design of **V-Prediction** models and avoid issues caused by incompatible samplers or configurations.

### 🧪 Inference Parameters

| Setting           | Recommended Value                        |
|-------------------|------------------------------------------|
| **CFG Scale**     | 4 ~ 5                                    |
| **Steps**         | 28 ~ 35                                  |
| **Sampler**       | `Euler` (⚠️ Other samplers may break)     |
| **Resolution**    | Target ~1024x1024 total area             |
|                   | Examples: `768x1344`, `832x1216`, `896x1152`, `1024x1024`, `1152x896`, `1216x832`, `1344x768` |

---

### 🧾 Prompt Structure

**Prompt Prefix (positive):**
```text
masterpiece, best quality, newest, absurdres, highres, safe,
````

**Negative Prompt:**

```text
nsfw, worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthro
```

---

## Prompting Guidelines

### 🏷️ Tag Categories

Structure your prompt using meaningful tag layers:

```
<1girl/1boy/...>, <character name>, <series>, <artist>, <special tags>, <general tags>, <composition/color tags>
```

### 🎯 Quality Tags

|Percentile|Tag|
|---|---|
|> 95th|`masterpiece`|
|85–95th|`best quality`|
|60–85th|`good quality`|
|30–60th|`normal quality`|
|< 30th|`worst quality`|

These are based on normalized image popularity and recency-weighted ratings.

### 🎨 Aesthetic Tags

- `very awa` — top 5% by aesthetic score (waifu-scorer, aesthetic-shadow-v2)
- `worst aesthetic` — bottom 5%, avoid unless stylistically intended

### 🕒 Date Tags

Use these to match art periods and styles:

| Year Range | Tag      |
| ---------- | -------- |
| 2005–2010  | `old`    |
| 2011–2014  | `early`  |
| 2014–2017  | `mid`    |
| 2018–2020  | `recent` |
| 2021–2024  | `newest` |

---
## Prompting Tip

> For dramatic or dark scenes:  
> Add `low key lighting`, `night`, `dark`, `moody`, `candlelight`, or `backlight` to enhance contrast and depth without breaking the model.

---
## Recipe NAI-X v2
```
version 0.1.0
model "NAI\\vpred\\noobai-vpred-kirazuri-lazuli-1.1.safetensors" model_config="sdxl-sgm" merge_space="weight"
model "NAI\\vpred\\kirazuriLazuliNoobaiV_v20NoobaiVPred1.safetensors" model_config="sdxl-sgm" merge_space="weight"
model "NAI\\vpred\\hikariNoobVPred_121.safetensors" model_config="sdxl-sgm" merge_space="weight"
model "NAI\\vpred\\MIDNIGHT_NAI-XL_vPredV1_Baked_VAE.safetensors" model_config="sdxl-sgm" merge_space="weight"
model "NAI\\vpred\\NoobAI-XL-Vpred-v1.0.safetensors" model_config="sdxl-sgm" merge_space="weight"
merge "karcher_mean" &0 &1 &2 &3 &4 max_iter=30.000000000000007 tol="3e-06" alphas="0.22, 0.12, 0.22, 0.22, 0.22"
```

> **Karcher mean** (Riemannian center of mass): a generalization of the arithmetic mean to curved spaces (Riemannian manifolds), finding the point that minimizes the sum of squared Riemannian distances to all checkpoints.

---
## Recipe NAI-X v1
***NAI-X-base**
```
model "[NAI]\NoobAI-XL-Vpred-v1.0.safetensors" "sdxl" "base"
model "[NAI]\hikariNoobVPred_121.safetensors" "sdxl" "base"
merge "subtract" &1 &0
model "[NAI]\waiSHUFFLENOOB_vPred20.safetensors" "sdxl" "base"
merge "subtract" &3 &0
merge "ties_sum" &2 &4 k=0.10000000000000002 vote_sgn=0.0
merge "add_difference" &0 &5 alpha=0.5000000000000001
model "[SDXL]\cyberrealisticXL_v53.safetensors" "sdxl" "base"
model "[SDXL]\sd_xl_base_1.0_0.9vae.safetensors" "sdxl" "base"
merge "subtract" &7 &8
merge "add_difference" &6 &9 alpha=1.0000000000000002
dict sdxl_txt2_default=0.0 sdxl_unet_default=1.0000000000000002 sdxl_txt_default=0.0
merge "clamp" &10 &0 &1 &3 stiffness=&11
```
**NAI-X-zero**
```
add lora DPO 
```
**NAI-X-soft**
```
A NoobAI-XL-Vpred-v1.0.safetensors
B NAI-X-zero

clip 
dare-ties(sum) + clerp
ratio 0.75 drop 0.25 iterations 3

u-net
MBW [1,1,0,0,0,0,0.2,0.4,0.6,0.8,1,1,1,1,1,1,1,1,1,1,1,1,1]
```