PixelNinjaArt_LoRA_Flux1 Repo update. Uploading the following: Sample images of training steps 0-2000 session. LoRA file. README.md, _latent_cache archive, aptimizer.pt archive, config..yaml file. Style accuracy is Noticeably better thayt previous versions I trained on civitai. Something I also did differently Regarding training in this session is that I removed a few images from the old dataset as well as made sure that the number of images was an even number. According to my own research, as well as my own experience, Having an odd number of images causes issues with memory management (think of the baking an unevern number of muffins analogy and why doing that would be ineffecient.). Due to the sheer amount of V Ram I had available for this, Doing this probably wasn`t totally necessary, but I feel it`s a good practice and habit to do. The original data set had 93 images, but I trimmed it down to eighty six this time. Anatomical degredation seem to be way less Prevalent when using this version. This current version is trained to 2000 steps, but I will likely train it up to 4000 in the future to see if I can get even better accuracy without having any anatomical degradation. In past training sessions on the on site Civitai trainer, having images that were even Remotely NSFW caused generating images to have the same style as the data said, but have severe anatomical degradation. extra limbs. bubbly looking lambs. missing limbs in general. etc etc. Well, it did happen a little bit with this version, it was nowhere near as bad as previous versions Perhaps I had different settings, or perhaps there is there`s something different going on when using the ostris ai toolkit trainer (https://github.com/ostris/ai-toolkit & https://github.com/AiArtFactory/ai-toolkit ). To make this very long commit message short, it looks promising

images/00000-386560115.png → D-ART-Flux1.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d83a81c41d762436060e5cf5e66d03507c603f2a54eefc338e4b33dd21dca681
-size 2036550
+oid sha256:c698479dbd71e571db6f98e9f2d1b600a21f4df68266d996fc0069b05ed8100b
+size 171969344

README.md

@@ -1,160 +1,49 @@
 ---
 tags:
 - text-to-image
-- stable-diffusion
 - lora
 - diffusers
-- template:sd-lora
+- template:diffusion-lora
+instance_prompt: D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1
+base_model: black-forest-labs/FLUX.1-dev
 widget:
 - text: >-
-    anime style picture, skinny and pretty  woman with white hair and "depicting
-    an extreme close up face of a dainty young mischievous female stoner prep
-    highschool school student with medium length silky straight iridescent black
-    hair and lightly suntanned skin,
-  output:
-    url: images/1000093792.jpeg
-- text: >-
-    motion lines, 1 cartoon girl,  angel with 4 large angel wings and a halo,
-    perfect curves, wearing a white dress, strapless, suggestive pose, romantic,
-    solo, detailed eyes, looking at viewer, sitting on a rock, forest, nature,
-    forest meadow, beautiful, looking at viewer, embarrassed, medium breasts,
-    from front, looking at viewer playfully, naughty, smirk, romantic, (blonde
-    hair, mid parted hair, straight hair, long hair:1.2), pronounced blush,
-    beautiful green eyes,<lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00024-2777017634.png
-- text: >-
-    motion lines, 1girl, dark skinned girl of African American decent, she had
-    red eyes and short black wavy hair, smiling at viewer, head shot, body not
-    visible, rosey red cheeks and blushing, white background, the woman has red
-    color accents in her hair, floating korean text is visible in the white
-    background <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00023-734285917.png
-- text: >-
-    motion lines, 1girl Tomboy, short purple hair, black t-shirt, midriff ,
-    navel ,smug smile, shimmering purple manga background with celestial stars,
-    looking directly at the viewer, <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00021-386560115.png
-- text: >-
-    1girl standing in front of a full length mirror taking a selfie, holding a
-    cellphone while wearing elegant lingerie,  messy bedroom can be seen in the
-    background, <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00019-2889018583.png
-- text: >-
-    highly detailed anime picture, a man in nursing scrubs stands in a hallway,
-    waist up portrait shot, beard and mustache, blurry background with a depth
-    of field effect, he is looking at the viewer and smiling, 
-    <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00016-522577074.png
-- text: >-
-    highly detailed anime picture, waist up portrait shot picture of a woman
-    laying on the floor inside of an indoors server room, above shot, computer
-    wires all over the floor, looks tired with bags under her eyes
-    <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00015-3110212036.png
-- text: >-
-    Highly detailed anime style picture, waist up portrait shot of a woman
-    wearing metal medieval knight armor, breast plates and metal shoulder pads,
-    blonde hair and black eyes, white background, holding a sign that says "need
-    more buzz" <lora:D-ART-18dart5[Flux]:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a woman with blonde hair, wearing a white off-the-shoulder bodysuit,  She has a neutral expression and is looking directly at the camera. The background is a plain, light grey wall.  The image has a high-quality, professional feel."
   output:
-    url: images/00014-1334532871.png
+    url: samples_0-2000/1735888308294__000002000_0.jpg
 - text: >-
-    Highly detailed anime style picture, shot of a field of sunflowers with
-    beautiful clouds during a sunset, no humans or subjects are present, no
-    people, butterflies are flying,  <lora:D-ART-18dart5[Flux]:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a woman with blonde hair, wearing a black lace bra, She has a small silver heart-shaped pendant necklace around her neck. She is looking directly at the camera with a neutral expression. The background is a plain, light-colored wall.  The image is high quality .
   output:
-    url: images/00011-1071964309.png
+    url: samples_0-2000/1735888326881__000002000_1.jpg
 - text: >-
-    Highly detailed anime style picture, waist up shot of a woman wearing a
-    green cloak with the hood over her head, orange hair and blue eyes,
-    background is a damp gloomy looking forest with thick fog
-    <lora:D-ART-18dart5[Flux]:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a  woman with long blonds hair, wearing a white shirt with a plunging neckline,  She has a serious expression and is looking directly at the camera. The background is a plain, textured grey wall. The lighting is natural, highlighting her features. The image has a high-quality, professional feel. The woman is the main subject, positioned in the center of the frame. 
   output:
-    url: images/00010-1455996471.png
+    url: samples_0-2000/1735888345491__000002000_2.jpg
 - text: >-
-    Highly detailed anime style picture. great lighting, intricate detail,
-    beautiful, stunning space marine, with aqua eyes wearing a choker,
-    skin-tight black body armor with pink accents, cleavage is cut out, stading
-    at the edge of a beach and the ocean, Perfect Composition, movie scenes,
-    highly detailed. lighting, sharp contrast <lora:D-ART-18dart5[Flux]:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a woman with blonde hair wearing a black turtleneck sweater and a gold necklace with a large pendant. She is sitting in the passenger seat of a car, smiling at the camera with her right hand resting on her ear. The background shows a clear blue sky and some greenery outside the car window.  The image is high quality and has a professional feel. 
   output:
-    url: images/00009-2654087682.png
+    url: samples_0-2000/1735888364101__000002000_3.jpg
 - text: >-
-    1 anime girl, solo, long_hair, looking_at_viewer, wide hips, two arms,
-    ((upper body)), chubby, blush, smile, detailed face+blue_eyes, indoor,
-    black_hair, gloves, dress, ribbon, bare_shoulders, very_long_hair,
-    purple_eyes, hair_ribbon, ponytail, elbow_gloves, belt, bracelet,
-    hand_on_hip, bell, christmas, red_gloves,
-    <lora:DART__18dart5_-_Style_Flux_-V3:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a woman with blonde hair, wearing a blue and black Starfleet uniform with a gold Star Fleet com-badge on the chest. She is standing in a well lit room.  The woman has a serious expression and is looking directly at the camera. The image has a high-quality, professional feel."
   output:
-    url: images/00006-633262202.png
+    url: samples_0-2000/1735888382704__000002000_4.jpg
 - text: >-
-    1girl holding a sign that says "Triple Penetration?", blank background,
-    portrait shot, smile, blush, <lora:DART__18dart5_-_Style_Flux_-V3:1>
+    D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of a woman with blonde hair,wearing a blue t-shirt with white polka dots, standing in a snowy forest with tall pine trees in the background. She has a neutral expression and is looking directly at the camera.  The image is high quality and has a professional feel.
   output:
-    url: images/00005-3728505413.png
-- text: >-
-    motion lines, 1girl Tomboy, short black hair, black t-shirt, midriff , hands
-    behind back, navel ,smug smile, shimmering manga background,
-    <lora:DART__18dart5_-_Style_Flux_-V3:1>
-  output:
-    url: images/00003-460431563.png
-- text: >-
-    blond haired anime girl standing in front of a brick wall , brick wall in
-    the background shiny blonde hair, <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00012-3359356420.png
-- text: >-
-    1girl holding a sign that says "Are you using a LoRA?", blank background,
-    portrait shot, smile, blush <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00002-2491779138.png
-- text: >-
-    1girl holding a sign that says "DEGEN", blank background, portrait shot,
-    smile, blush <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00007-3864689592.png
-- text: >-
-    1girl, blonde hair, bob cut, fennec ears, purple oni horns, red eyeshadow,
-    purple eyes, purple lips, large breasts, kimono, (multiple tails, fox tail),
-    upper body, park,, <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00001-3338387262.png
-- text: >-
-    motion lines, 1girl Tomboy, short black hair, black t-shirt, midriff , navel
-    ,smug smile, shimmering manga background, <lora:D-ART-18dart5[Flux]:1>
-  output:
-    url: images/00000-386560115.png
-base_model: black-forest-labs/FLUX.1-dev
-instance_prompt: null
-license: other
-license_name: flux-1-dev-non-commercial-license
-license_link: https://huggingface.co/black-forest-labs/FLUX.1-dev/blob/main/LICENSE.md
-pipeline_tag: text-to-image
-datasets:
-- AiAF/D-ART_LoRA_Flux1_Dataset
+    url: samples_0-2000/1735888401314__000002000_5.jpg
 ---
-# D-ART | @18dart5 - Style LoRA [Flux1]
 
 <Gallery />
 
-## Model description 
+# D-ART | @18DART5 Style LoRA [Flux1]
 
-Flux1 Style LoRA trained on the artist @18dart5. No specific trigger words, but words like &quot;anime&quot;, &quot;cartoon&quot;, &quot;illustration&quot;, etc, can help keep in on track
+## Model description
 
- Please let me know your thoughts!
+LoRA based on the Artist, DART's ( https://x.com/18dart5 ) , artstyle.
 
-PonyXL version can also be downloaded here from Huggingface!
+Please let me know your thoughts!
 
-Huggingface: https:&#x2F;&#x2F;huggingface.co&#x2F;AiAF
-
-Twitter: AiArt Factory (@AiArtFactory) &#x2F; X (twitter.com)
+Twitter: https://x.com/AiArtFactory
 
 TikTok: https:&#x2F;&#x2F;www.tiktok.com&#x2F;@ai_art_factory?_t&#x3D;8cykD6v1PRu&amp;_r&#x3D;1
 
@@ -162,9 +51,6 @@ Instagram: https:&#x2F;&#x2F;instagram.com&#x2F;aiart.factory?igshid&#x3D;NGExMm
 
 Deviantart: https:&#x2F;&#x2F;www.deviantart.com&#x2F;aiartfactory
 
+## Trigger words
 
-## Download model
-
-Weights for this model are available in Safetensors format.
-
-[Download](/AiAF/D-ART_LoRA_Flux1/tree/main) them in the Files & versions tab.
+You should use `D-ART \(Artist\), @18dart5, @18dart3, @18dart2,` and/or  `@18dart1` to trigger the image generation.

D-ART-18dart5[Flux1].safetensors → _latent_cache.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84d3c509a8e58132aff7fe3ddcb3b9f9e10147f32c1af38de14bab62d09ad2fe
-size 19299736
+oid sha256:193bc31b10fd23c2aad175317f84bba978ad35fb7cce8bb9abe4724786fb0d13
+size 53063198

config.yaml

@@ -0,0 +1,89 @@
+job: extension
+config:
+  name: D-ART-Flux1
+  process:
+  - type: sd_trainer
+    training_folder: /content/output
+    device: cuda:0
+    network:
+      type: lora
+      linear: 16
+      linear_alpha: 16
+    save:
+      dtype: float16
+      save_every: 250
+      max_step_saves_to_keep: 10
+    datasets:
+    - folder_path: /content/dataset
+      caption_ext: txt
+      caption_dropout_rate: 0.05
+      shuffle_tokens: false
+      cache_latents_to_disk: true
+      resolution:
+      - 512
+      - 768
+      - 1024
+    train:
+      batch_size: 1
+      steps: 2000
+      gradient_accumulation_steps: 1
+      train_unet: true
+      train_text_encoder: false
+      content_or_style: content
+      gradient_checkpointing: true
+      noise_scheduler: flowmatch
+      optimizer: adamw8bit
+      lr: 0.0004
+      ema_config:
+        use_ema: true
+        ema_decay: 0.99
+      dtype: bf16
+    model:
+      name_or_path: black-forest-labs/FLUX.1-dev
+      is_flux: true
+      quantize: true
+    sample:
+      sampler: flowmatch
+      sample_every: 250
+      width: 1024
+      height: 1024
+      prompts:
+      - D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a woman with blonde hair, wearing a white off-the-shoulder bodysuit,  She
+        has a neutral expression and is looking directly at the camera. The background
+        is a plain, light grey wall.  The image has a high-quality, professional feel.
+      - D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a woman with blonde hair, wearing a black lace bra, She has a small silver
+        heart-shaped pendant necklace around her neck. She is looking directly at
+        the camera with a neutral expression. The background is a plain, light-colored
+        wall.  The image is high quality .
+      - 'D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a  woman with long blonds hair, wearing a white shirt with a plunging neckline,  She
+        has a serious expression and is looking directly at the camera. The background
+        is a plain, textured grey wall. The lighting is natural, highlighting her
+        features. The image has a high-quality, professional feel. The woman is the
+        main subject, positioned in the center of the frame. '
+      - 'D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a woman with blonde hair wearing a black turtleneck sweater and a gold necklace
+        with a large pendant. She is sitting in the passenger seat of a car, smiling
+        at the camera with her right hand resting on her ear. The background shows
+        a clear blue sky and some greenery outside the car window.  The image is high
+        quality and has a professional feel. '
+      - D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a woman with blonde hair, wearing a blue and black Starfleet uniform with
+        a gold Star Fleet com-badge on the chest. She is standing in a well lit room.  The
+        woman has a serious expression and is looking directly at the camera. The
+        image has a high-quality, professional feel.
+      - D-ART \(Artist\), @18dart5, @18dart3, @18dart2, @18dart1, illustration of
+        a woman with blonde hair,wearing a blue t-shirt with white polka dots, standing
+        in a snowy forest with tall pine trees in the background. She has a neutral
+        expression and is looking directly at the camera.  The image is high quality
+        and has a professional feel.
+      neg: ''
+      seed: 9000
+      walk_seed: true
+      guidance_scale: 4
+      sample_steps: 20
+meta:
+  name: D-ART-Flux1
+  version: '3.0'

images/00001-3338387262.png → optimizer.pt.zip

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e2a9ba05169e79c9282ef196704d37e11659d622ccb1e5b0f18e4c43c7e483d5
-size 1143856
+oid sha256:263bcd8624a2a697435e9ebf771e0fcd5ed033fb348b37730bb6c981358bf8f0
+size 155119465