Update README.md

README.md

@@ -106,7 +106,7 @@ base_model: cagliostrolab/animagine-xl-3.1
  }
 
  .overlay-text {
-   background: linear-gradient(45deg, #7ed56f, #28b485);
+   background: linear-gradient(45deg, #F1F8E8, #F1F8E8);
    -webkit-background-clip: text;
    color: transparent; 
    text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.7); 
@@ -132,25 +132,25 @@ base_model: cagliostrolab/animagine-xl-3.1
  <tr>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 1">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/LrSWVz5LpvRPmYjzZxm_F.png" alt="Sample Image 1">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 1</div>
+         <div class="overlay-text">1girl, satanichia kurumizawa mcdowell, gabriel dropout, fang, collared shirt, red hair, red eyes, hair rings, bat hair ornament, school uniform, red necktie, black shirt, pleated skirt, red skirt, smile, open mouth, orange cardigan, v, blush, indoors, classroom,</div>
        </div>
      </div>
    </td>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 2">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/T-4YVVNz8S2vqzI-Zivnb.png" alt="Sample Image 2">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 2</div>
+         <div class="overlay-text">1girl, erza scarlet, fairy tail, red hair, brown eyes, armor, blue skirt, breastplate, pauldrons, gauntlets, shoulder armor, boots, sword, weapon, holding sword, holding weapon,</div>
        </div>
      </div>
    </td>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 3">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/p3iw5gkHHC1FbgnINDD2A.png" alt="Sample Image 3">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 3</div>
+         <div class="overlay-text">1girl, hinata kaho, blend s, blonde hair, blue eyes, twintails, long hair, head scarf, stile uniform, head scarf, blue shirt, frills, frilled apron, collared shirt, puffy sleeves, puffy short sleeves, medium breasts, waist apron, white gloves, white apron, cup, coffee, tray, smile, wavy mouth, open mouth,</div>
        </div>
      </div>
    </td>
@@ -158,25 +158,25 @@ base_model: cagliostrolab/animagine-xl-3.1
  <tr>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 4">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/qHZ8HMT8PnPG-9QDP92Og.png" alt="Sample Image 4">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 4</div>
+         <div class="overlay-text">2girls, kirisaki chitoge, onodera kosaki, nisekoi, looking at viewer, side-by-side, standing, school uniform, serafuku, white shirt, blue skirt, v,</div>
        </div>
      </div>
    </td>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 5">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/tLozu285gpCwoRBGlHYDe.png" alt="Sample Image 5">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 5</div>
+         <div class="overlay-text">1girl, katou megumi, saenai heroine no sodatekata, brown hair, brown eyes, collared dress, white dress, ribbon, jacket, beret, smile, looking at viewer, outdoors, flower field, sky,</div>
        </div>
      </div>
    </td>
    <td>
      <div class="custom-image-container">
-       <img class="custom-image" src="" alt="Sample Image 6">
+       <img class="custom-image" src="https://cdn-uploads.huggingface.co/production/uploads/64b24543eec33e27dc9a6eca/MsDanEXq2z9J_P7e5zGtf.png" alt="Sample Image 6">
        <div class="overlay">
-         <div class="overlay-text">Sample Image 6</div>
+         <div class="overlay-text">1girl, jabami yumeko, kakegurui, black hair, red eyes, long hair, blunt bangs, hyakkaou academy school uniform, school uniform, collared shirt, white shirt, ribbon, red jacket, pleated skirt, print skirt, pantyhose, smile, open mouth, card, playing card,</div>
        </div>
      </div>
    </td>
@@ -184,12 +184,12 @@ base_model: cagliostrolab/animagine-xl-3.1
 </table>
 
 ## Overview
-Introducing the enhanced **Rae Diffusion XL** , an advanced iteration of the Animagine XL 3.1 model, specifically fine-tuned for generating stunning anime-style artwork. Building on the robust foundation of the SDXL model, Rae Diffusion XL is meticulously optimized to excel in depicting anime characters, pushing the boundaries of creativity and technical capability.
+Introducing **Rae Diffusion XL** , an enhanced iteration based on the Animagine XL 3.1 model, specifically fine-tuned for generating stunning anime-style artwork. Building on the robust foundation of the SDXL model, **Rae Diffusion XL** is meticulously optimized to excel in depicting anime characters, pushing the boundaries of creativity.
 
 ## Model Details
 - **Developed by**: [Raelina](https://civitai.com/user/Raelina)
 - **Model type**: Diffusion-based text-to-image generative model
-- **Model Description**: Generate high-quality anime images from textual prompts
+- **Model Description**: Rae Diffusion XL is an enhanced iteration built on the Animagine XL 3.1 model. It is fine-tuned for high-quality anime-style character art generation.
 - **License**: [Fair AI Public License 1.0-SD](https://freedevproject.org/faipl-1.0-sd/)
 - **Finetuned from**: [Animagine XL 3.1](https://huggingface.co/cagliostrolab/animagine-xl-3.1)
 
@@ -211,16 +211,17 @@ Rae Diffusion XL inherits special tags from Animagine XL 3.1 to enhance image ge
   
 ## Recommended settings
 
-- **Negative prompts:**
+- **Positive prompts:**
+
 ```
-(worst quality, low quality, very displeasing), (interlocked fingers, badly drawn hands and fingers, anatomically incorrect hands), signature, watermark, artist name,
+masterpiece, best quality, ultra detailed, very aesthetic,
 ```
 
-- **Positive prompts:**
-
+- **Negative prompts:**
 ```
-masterpiece, best quality, very aesthetic, intricate details,
+(low quality, worst quality:1.2), very displeasing, ugly, interlocked fingers, badly drawn, anatomically incorrect hands, signature, watermark,
 ```
+
 - **CFG:** 7
 - **Sampling steps:** 25 to 30
 - **Sampler:** Euler a
@@ -234,19 +235,18 @@ masterpiece, best quality, very aesthetic, intricate details,
 - **Hires step:** 10-15
 - **Denoising:** 0.1-0.3 or 0.55 for latent upscaler
 
-## Training parameters
+## Training config
 - Hardware: 1x RTX 3090 24gb
 - Batch size: 12
-- Gradient Accumulation: 3
-- Epochs: 10
-- Unet LR: 3e-04
-- Text Encoder LR: 1e-04
-- Optimizer: AdamW8bit
-- Optimizer Args: Weight Decay: 0.1, Betas: (0.9, 0.99)
-- LR Scheduler: Cousine
-- Min SNR Gamma: 5
-- Algorithm: LoKr
-- Total Training Time: 46 hours
+- Gradient Accumulation: 1
+- Epochs: 15
+- Learning Rate: 5e-5
+- Optimizer: Adafactor
+- Optimizer Args: (Scale Parameter: False, Relative Step: False, Warmup Init: False)
+- Scheduler: Constant with warmup
+- Warmup steps: 0.05
+- Noise offset: 0.0357
+- Algorithm: Locon
 
 ## License
 
@@ -255,5 +255,3 @@ Rae Diffusion XL now uses the [Fair AI Public License 1.0-SD](https://freedevpro
 2. **Source Code Accessibility:** If your modified version is network-accessible, provide a way (like a download link) for others to get the source code. This applies to derived models too.
 3. **Distribution Terms:** Any distribution must be under this license or another with similar rules.
 4. **Compliance:** Non-compliance must be fixed within 30 days to avoid license termination, emphasizing transparency and adherence to open-source values.
-
-The choice of this license aims to keep Raemu XL open and modifiable, aligning with open source community spirit. It protects contributors and users, encouraging a collaborative, ethical open-source community. This ensures the model not only benefits from communal input but also respects open-source development freedoms.